
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                  BIOPESTICIDES REGISTRATION ACTION DOCUMENT
                                             
                                             
                                             

                        PLANT-INCORPORATED PROTECTANTS:
                                       
        Bacillus thuringiensis Cry1Ab protein and the Genetic Material
    Necessary for its Production in Event T304-40 Cotton [PC Code 006525, 
                    OECD Unique Identifier: BCS-GHØØ4-7];
        Bacillus thuringiensis Cry2Ae protein and the Genetic Material
     Necessary for its Production in Event GHB119 Cotton [PC Code 006600, 
                      OECD Unique Identifier: BCS-GH5-8];
                                      and
       TwinLink(TM) Cotton (T304-40 and GHB119 Combination PIP product)  
              [OECD Unique Identifier: BCS-GHØØ4-7 x BCS-GH5-8]
                                       
                                       
                                       
                                       
                                       
                                       
                                       
                     U.S. Environmental Protection Agency
                         Office of Pesticide Programs
                Biopesticides and Pollution Prevention Division
                                       
                                       
                               January 26, 2012

                  
    Bacillus thuringiensis Cry1Ab, Cry2Ae, TwinLink(TM) Cotton (Cry1Abx2Ae) 
                               Table of Contents
BIOPESTICIDES REGISTRATION ACTION DOCUMENT TEAM	3
GLOSSARY OF ACRONYMS AND ABBREVIATIONS	4
EXECUTIVE SUMMARY	5
I.	ACTIVE INGREDIENT OVERVIEW	6
II.    REGULATORY BACKGROUND	7
III.   RISK ASSESSMENT SUMMARIES	10
A.	Product Characterization	10
B.	Human Health Assessment	11
C.   Environmental Effects Assessment	17
D.   Insect Resistance Management	18
IV.	ENVIRONMENTAL JUSTICE	22
V. 	RISK MANAGEMENT AND PROPOSED REGISTRATION DECISION	23
VI.      TERMS AND CONDITIONS OF THE REGISTRATION(S)	23
APPENDIX A:  Product characterization and human health effects  -  Bacillus thuringiensis Cry 1Ab (Event T304-40) in cotton plant incorporated protectant.	26
APPENDIX B:  Product characterization and human health effects - Bacillus thuringiensis Cry 2Ae (Event GHB 119) and TwinLink(TM) in cotton plant incorporated protectant.	36
APPENDIX C: Ecological and environmental effects  -  Bacillus thuringiensis Cry1Ab, Cry2Ae and TwinLink(TM) cotton plant incorporated protectant	58
APPENDIX D: Insect Resistance Management	85
APPENDIX E: References	134






BIOPESTICIDES REGISTRATION ACTION DOCUMENT TEAM
                                       
                          Microbial Pesticides Branch
                                       
Product Characterization and Human Health
John Kough, Ph.D.
Annabel Waggoner, B.S.
Chris Wozniak, Ph.D.

Environmental Fate and Effects
Zigfridas Vaituzis, Ph.D.
Shannon Borges, M.S.
Annabel Waggoner, B.S.

Insect Resistance Management
Jeannette Martinez, M.S.
Alan Reynolds, M.S.

Benefit Assessment
Shannon Borges, M.S.

Office of General Counsel
Chris Kaczmarek

Biopesticides Registration Action Document Regulations
Sheryl Reilly, Ph.D. 
Alan Reynolds, M.S. 
Shanaz Bacchus, M.S., MBA
Denise Greenway, B.S.

Previous U.S. EPA Reviewers
Rebbeca Edelstein, Ph.D.
Sharlene Matten, Ph.D.

				

USDA/BRS Work share Program
Patricia Beetham, Ph.D. USDA/BRS



GLOSSARY OF ACRONYMS AND ABBREVIATIONS 

APHIS			Animal and Plant Health Inspection Service (USDA)
BPPD			Biopesticides and Pollution Prevention Division
BRAD			Biopesticides Registration Action Document
b.w.			body weight
CAS			Chemical Abstracts Service
40 CFR			Title 40 of the Code of Federal Regulations
cDNA			Copied (or Copy) DNA
cfu			colony forming units
dsRNA			Double-Stranded RNA
cos			Cosmid
[o]C			Temperature in Centigrade or Celsius Degrees
DER			Data Evaluation Record
DNA			Deoxyribonucleic Acid
EPA			Environmental Protection Agency (the "Agency")
ELISA			Enzyme-Linked Immunosorbent Assay
EPA Reg. No.		EPA Registration Number
FFDCA			Federal Food, Drug, and Cosmetic Act
FIFRA			Federal Insecticide, Fungicide, and Rodenticide Act
FQPA 			Food Quality Protection Act
FR			Federal Register
g 			Gram
IU			International Units			
IRM			Insect Resistance Management
kg			kilogram
L			Liter
MRID No.		Master Record Identification Number
mg			Milligram
mL			Milliliter
μg			Microgram
MP			Manufacturing-Use Product
mRNA			Messenger RNA
NE			No Effect
ng			nanogram (10[-9] gram)
NIOSH 			National Institute for Occupational Safety and Health
OPP			Office of Pesticide Programs
OCSPP			Office of Chemical Safety and Pollution Prevention
OECD			Organisation for Economic and Community Development
PC 			Pesticide Chemical
PCR			Polymerase Chain Reaction 
PIP			Plant-Incorporated Protectant
pg			picogram (10[-12] gram)
ppb			parts per trillion
ppm			parts per million
ppt			parts per trillion
PTGS			Post-Transcriptional Gene Silencing
RNA			Ribonucleic Acid
rRNA			Ribosomal RNA
T-DNA 			transfer DNA from Agrobacterium
TGAI			Technical Grade of the Active Ingredient
USDA			United States Department of Agriculture
U.S. EPA		United States Environmental Protection Agency
U.S. FDA		United States Food and Drug Administration



EXECUTIVE SUMMARY

The United States Environmental Protection Agency (U.S. EPA, Agency) proposes to register three new cotton plant-incorporated protectant (PIP) products as described in this Biopesticide Registration Action Document (BRAD).These registration actions represent one new PIP active ingredient and a new event of a PIP that has been previously registered in corn. The expression of the PIPs in these products represent their first food and first outdoor uses in cotton. 

The Agency established a policy in October 1, 2009 to inform and provide the public an opportunity to comment on proposed registration decisions before they occur. The draft BRAD and supporting documents were available for 30 days in the Agency's docket (EPA-HQ-OPP-2011-0988) for this purpose. No comments were received from the public at the end of the public comment period on January 13, 2012.

Bayer CropScience, LP (BCS) developed and sought federal registration of two transgenic cotton (Gossypium hirsutum) PIPs, Event T304-40 and Event GHB119. Each expresses an insecticidal protein (also known as "Cry" proteins) derived from the soil bacterium Bacillus thuringiensis (Bt). These cotton events are intended for use as PIPs for conventional breeding to produce a product, TwinLink(TM) Cotton (referred to in this document as "TwinLink(TM)") that expresses both proteins. 

This BRAD presents EPA's evaluations of the data and information submitted by BCS for the proposed registrations of: 

   1. EPA File Symbol 264-RNOU: Event T304-40 cotton expressing Bt Cry1Ab insecticidal protein. The cry1Ab gene that expresses this protein is derived from Bt subspecies berliner. It was created by Agrobacterium-mediated transformation of Gossypium hirsutum cv. Coker with the transfer DNA (TDNA) vector pTDL008. 
         
   2. EPA File Symbol 264-RNOL: Event GHB119 cotton expressing Bt Cry2Ae insecticidal protein. It was created by Agrobacterium-mediated transformation using vector pTEM12. It contains the cry2ae gene isolated from Bt subspecies dakota.

   3. EPA File Symbol 264-RNOA: Genetic material from Events T304-40 and GHB119 are combined by conventional breeding to express the combined proteins, Bt Cry1Ab and Bt Cry2Ae, in TwinLink(TM) Cotton. 

The genes that are responsible for the Bt Cry1Ab and Bt Cry2Ae proteins, cry1Ab and cry2Ae, have been modified for their expression in plants. These proteins are referred to as Cry1Ab (or Event T304-40), and Cry2Ae (or Event GHB 119), respectively, in this BRAD. Both events also include the pat gene that produces Phosphinothricin Acetyltransferase (PAT) enzyme, an inert ingredient in the proposed products that confers tolerance of the cotton plant to the herbicide, glufosinate. 

An Experimental Use Permit (EUP) was issued to BCS to conduct field trials with Event T304-40 (Cry1Ab) cotton on February 7, 2006 and was extended to expire on December 31, 2011. An EUP for GHB119 (Cry2Ae) and the combination PIP product, TwinLink(TM) (expressing Cry1Ab and Cry2Ae proteins) was issued on September 1, 2008, and will expire on December 31, 2011. These EUPs were associated with permanent tolerance exemptions for the active ingredient, Cry1Ab protein (40 CFR § 174. 511), and the inert enzyme, PAT protein (40 CFR § 174. 522), and a temporary tolerance exemption for the active ingredient, Cry2Ae protein (40 CFR § 174. 530).

BCS applied for pesticide registrations of Cry1Ab and Cry2Ae proteins as expressed in Events T304-40 and GHB119 cotton, respectively, and a new combination PIP product, TwinLink(TM) cotton, created through conventional breeding of these events for use as PIPs in 2009, under section 3 of the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA). BCS concurrently filed a petition for a permanent exemption from the requirement of a tolerance for Cry2Ae protein residues in or on food and feed commodities in cotton. The Agency believes that, based upon its assessment of the data and information submitted by BCS, it is in the interest of the public and the environment to issue the registrations proposed by BCS.
I. ACTIVE INGREDIENT OVERVIEW

Active Ingredients
and Office of Pesticide Programs (OPP) Chemical Code and File Symbol:
  a. Bt Cry1Ab insecticidal protein and the genetic material necessary for its production in Event T304-40 cotton (PC Code 006525);  OECD Unique Identifier: BCS-GHØØ4-7;
     EPA File Symbol 264-RNOU.

  b. Bt Cry2Ae insecticidal protein and the genetic material necessary for its production in Event GHB119 cotton (PC Code 006600); OECD Unique identifier: BCS-GH5-8; 
        EPA File Symbol 264-RNOL.
  
  c. Bt Cry1Ab and Bt Cry2Ae insecticidal proteins and the genetic material necessary for their production in TwinLink(TM) cotton (T304-40 x GHB119) combination PIP product; OECD Unique Identifier: BCS-GHØØ4-7 x BCS-GH5-8;
     EPA File Symbol 264-RNOA.
     
Applicant/
Manufacturer:
Bayer CropScience  -  BioScience (BCS)
P.O. Box 12014
2 T.W. Alexander Drive
Research Triangle Park, NC 27709

Type of Pesticide:
Plant Incorporated Protectant (PIP)

Use:
Insecticide

Target Pest:
Cotton bollworm (CBW, Helicoverpa zea), pink bollworm (PBW, Pectinophora gossypiella), tobacco budworm (TBW, Heliothis virenscens) larvae, beet armyworm (Spodoptera exigua), and fall armyworm (Spodoptera frugiperda). 

Mode of action:

Insecticides formulated with microbial Bt have been in use worldwide since 1961. These microbial active ingredients are the source of the genes that express the Cry1Ab and Cry2Ae proteins in the proposed PIPs. There is no mechanistic evidence that Bt proteins pose risks for humans or mammals during their use as pesticides or as PIPs (OECD, 2007). When Cry proteins are ingested by an insect, they are dissolved under the alkaline conditions specific to the insect's gut, and then activated upon cleavage by the insect's midgut enzymes (proteases). These activated Cry proteins must bind specifically to glycoprotein or glycolipid receptors on microvillar membrane of the insect's midgut to initiate the pore-forming process that ultimately kills the insect. The fragment of the protein that is toxic to insects then interacts with specific high-affinity receptors on the microvilli of the target insect's midgut epithelium (stomach), in particular brush border membrane vesicles (BBMV) (Ferré and Van Rie, 2002). Such receptors are not present in vertebrate species. Cry proteins must enter the cell membrane and form a pore. 

Although receptor binding appears to be essential for the insecticidal activity of the crystal proteins, binding by itself may not lead to toxicity. It appears that the initial protein-receptor interaction or binding is a reversible process, while the irreversible insertion of at least part of the Cry protein (the fragment that is toxic to insects) into the cell membrane is responsible for forming the pore. Consequently, the ion gradient across the membrane is compromised, and the columnar cells of the midgut swell and lyse osmotically. Lysis leads to disruption of the gut epithelium, resulting in starvation, and insect death (Ferré and Van Rie, 2002). 
I. REGULATORY BACKGROUND  

Applications for use of Cry1Ab, Cry2Ae and TwinLink(TM) PIPs

   A. Cry 1Ab protein

   1. Experimental Use Permit and Temporary Tolerance Exemption 
            
         a. Cry1Ab cotton Permit (EPA Reg. No. 264-EUP-140)

EPA granted an Experimental Use Permit (EUP) for the Cry 1Ab cotton PIP to BCS on February 7, 2006 (71 FR 41020; July 19, 2006; FRL-8060-6). These experimental trials were conducted with cotton PIPs containing the genetic material necessary for Cry1Ab protein production in Events T303-3 and T304-40. The company intended to research the potential of this cotton PIP, Cry1Ab protein, for control of lepidopteran larvae, such as cotton bollworm (Helicoverpa zea) and tobacco budworm (Heliothis virescens), which are common pests of cotton. The EUP was amended on March 8, 2007 (72 FR 34009; June 20, 2007; FRL-8133-5), on August 28, 2008 (73 FR 58949; October 8, 2008; FRL-8384-9), and on November 24, 2008 (74 FR 10571; March 11, 2009; FRL-8398-2), to permit planting until December 31, 2010. An additional extension was granted to extend the amended program to run from January 1, 2011 until December 31, 2011. 

The proposed program was conducted in the States of Alabama, Arkansas, Arizona, California, Florida, Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee, Texas, and in the United States territory of Puerto Rico. The EUP included trials to evaluate insect and herbicide efficacy, agronomic performance, and breeding lines. Also, seed for future plantings of experimental and regulatory field trials were produced.
   


         b. Cry1Ab Temporary Tolerance Exemption 

A temporary tolerance exemption for Cry1Ab was not required, since a permanent tolerance exemption exists in 40 CFR § 174.511 for Cry1Ab protein in all food and animal feed commodities (discussed below in IIA.2.b.).
   2. Registration Application and Permanent Tolerance Exemption
            a.             Cry1Ab PIP FIFRA Section 3 Registration Application 
On April 8, 2009, EPA published a Notice of Receipt in the Federal Register (74 FR 15978), announcing that BCS submitted an application to register a new active ingredient, Bt Cry1Ab insecticidal protein (Event T304-40; EPA File Symbol 264-RNOU), not included in any currently registered pesticide products for use in cotton. No public comments were received in response to this publication.
            b.             Cry1Ab protein - Permanent Tolerance exemption  
Residues of Cry1Ab protein are exempt from the requirement of a tolerance when used as PIPs in all food commodities (40 CFR § 174.511). EPA granted an exemption from tolerance for Cry1Ab protein in all food and feed commodities on August 2, 1996. The tolerance exemption is published in the Code of Federal Regulations (40 CFR §180.1173). In September 2001, EPA completed a reassessment of this tolerance exemption considering all of the existing data, public literature, and public comments. The reassessment determined that the tolerance exemption met all the required scientific and regulatory standards. This tolerance exemption for the Cry1Ab protein is not event-specific and applies to all Cry1Ab protein and any other event producing the Cry1Ab protein that might be found in the food supply. On April 25, 2007, the tolerance exemption for Cry1Ab protein was reassigned to 40 CFR § 174.511 (72 FR 20435). Based on this existing tolerance exemption, a petition to establish a tolerance or tolerance exemption was not required for the Cry1Ab protein  expressed in this cotton PIP. 


   A. Bt Cry2Ae and TwinLink(TM) PIPs
    
      1. Experimental Use Permit and Temporary Tolerance Exemption 

            a.             Cry2Ae and TwinLink(TM) Cotton PIPs Experimental Use Permit 
            
In addition to the EUP for Cry1Ab cotton PIP, EPA granted an EUP to BCS for the Cry2Ae cotton PIP and the combined traits of Cry 1Ab x Cry2Ae proteins in TwinLink(TM) Cotton (EPA Reg. No. 264-EUP-143) on September 1, 2008 (73 FR 65848; November 5, 2008; FRL-8388-6), with an expiration date of  December 31, 2010. The permit was extended on December 30, 2010 to allow further evaluation of these cotton plant lines in a wider range of environmental conditions between January 1, 2011 and December 31, 2011. Testing was intended to include insect efficacy trials, agronomic performance evaluations, and herbicide efficacy evaluations, as well as the production of sample material for conducting regulatory studies. Tests were allowed in 12 states: Alabama, Arkansas, Arizona, California, Florida, Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee and Texas and the United States territory of Puerto Rico. Testing was authorized on a total of 1,919 acres of which 152 acres were to be planted to Cry2Ae parental cotton and 307 acres to Cry1Ab x Cry2Ae combined trait TwinLink(TM) Cotton. 

The company was researching the potential for Cry2Ae PIP and combined Cry1Ab and Cry2Ae proteins (TwinLink(TM) Cotton) produced by the inserted genetic material in these cotton PIPs for control of lepidopteran larvae, such as Cotton bollworm (Helicoverpa zea), Tobacco budworm (Heliothis virescens), Pink bollworm (Pectinophora gossypiella), Fall armyworm (Spodoptera frugiperda), and Beet armyworm (Spodoptera exigua)  All cotton plants evaluated under this EUP contained the Cry2Ae protein, and were derived from either a transformation Event (GHB119 or GHB714) or combinations derived from either transformation Event T303-3 or T304-40 (cry1Ab) and Event GHB119 or GHB714 (cry2Ae).

            b.             Cry2Ae Protein: Temporary Tolerance Exemption 
A notice of filing of a pesticide petition (PP7F7912) by BCS was published in the Federal Register of August 8, 2007 (72 FR 44521- 44523) (FRL-8139-7) in connection with the above referenced Cry2Ae PIP EUP. Relevant company data/information in support of the application was posted in the docket, EPA-HQ-OPP-2007-0573. Following EPA's evaluation of these submissions, a temporary exemption from tolerance was approved for residues of Bt Cry2Ae protein in/on cotton food/feed commodities (73 FR 52591; 9/10/2008). It will expire on December 31, 2012.
2. Registration Application and Permanent Tolerance Exemption
         a. Cry2Ae and TwinLink(TM) Cotton PIPs - FIFRA section 3 application 
On April 8, 2009, EPA published a Notice of Receipt (NOR) in the Federal Register (74 FR 15978), announcing that BCS submitted an application to register a new active ingredient, Bacillus thuringiensis Cry2Ae protein (EPA File Symbol 264-RNOL; Event GHB 119) as a PIP in cotton. BCS also proposed the registration of another new PIP product, Twinlink(TM) Cotton (EPA File Symbol 264-RNOA), developed by conventional breeding of the aforesaid cotton Events GHB119 and T304-40, thus expressing the combined traits of the two new active ingredients, Cry1Ab, and Cry2Ae proteins.
         b. Cry2Ae PIP  -  Permanent Tolerance Exemption

Concurrently with the FIFRA section applications, BCS submitted a petition (PP9F7514) establish a permanent exemption from the requirement of a tolerance for residues of Bacillus thuringiensis Cry2Ae protein in or on all food commodities. EPA published a Notice of Filing of the petition in the Federal Register on April 8, 2009 (74 FR 15970), and the public was given a 30-day comment period. 

Event GHB119 cotton also expresses the PAT enzyme that is exempt from the requirement of a tolerance when used as a PIP inert ingredient in all food commodities (40 CFR §174.522).
       
No public comments were received to any of the Federal Register announcements for Cry2Ae or TwinLink(TM) Cotton PIPs, or the notice of filing of the pesticide petition for Cry2Ae protein published in the Federal Register.

III.	RISK ASSESSMENT SUMMARIES

EPA assessed the data submitted by BCS in 2009 for the proposed PIP cotton registrations and is presenting summaries of the product characterization and human health risk assessments in Appendices A (Cry1Ab) and B (Cry 2Ae and TwinLink(TM)). Summaries of ecological and environmental risk assessments and Insect Resistance Management (IRM) for all three PIP products are presented in Appendices C and D, respectively. In its assessment, the EPA relied upon data and other information submitted by the applicant and general knowledge of best available scientific technology. 

The classifications that are found for each data submission are assigned by Agency science reviewers and are an indication of the usefulness of the information contained in the documents for risk assessment. A rating of "ACCEPTABLE" indicates the study is scientifically sound and is useful for risk assessment. A "SUPPLEMENTAL" rating indicates the data provide some information that can be useful for risk assessment. The studies may have certain aspects determined not to be scientifically acceptable ("SUPPLEMENTAL: UPGRADABLE"). If a study is rated as "SUPPLEMENTAL: UPGRADABLE," the Agency always provides an indication of what is lacking or what can be provided to change the rating to "ACCEPTABLE." If there is simply a "SUPPLEMENTAL" rating, the reviewer will often state that the study is not required by the current 40 Code of Federal Regulations (CFR) part 158. Both "ACCEPTABLE" and "SUPPLEMENTAL" studies may be used in the risk assessment process as appropriate. An "UNACCEPTABLE" rating indicates that new data need to be submitted.

For the acute toxicity data requirements, toxicity categories are assigned based on the 
hazard(s) identified from studies and/or other information submitted to the Agency in support of a pesticide registration. The active ingredient or particular product is classified into Toxicity Category I, II, III, or IV, where Toxicity Category I indicates the highest toxicity and Toxicity Category IV indicates the lowest toxicity. 

   A. Product Characterization 

      All product characterization data requirements for the Bt cotton PIPs containing Cry1Ab, Cry2Ae and their combined traits in TwinLink(TM) have been satisfied. The two cotton events developed by BCS to express these PIPs are:
      
         1. T304-40 cotton: expresses Cry1Ab insecticidal protein derived from Bt subspecies berliner. It was created by Agrobacterium-mediated transformation of Gossypium hirsutum cv. Coker with the TDNA vector pTDL008. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analyses of Cry1Ab protein isolated from Event T304-40 cotton leaves showed a very faint band estimated to be at a similar molecular weight as the band exhibited for the E. coli-derived Cry1Ab protein standard at approximately 66 kilodaltons (kDa). 
            
         2. Event GHB119 cotton: expresses Cry2Ae protein, derived from Bt subspecies dakota. It was created by Agrobacterium-mediated transformation using vector pTEM12,   and contains the cry2ae gene isolated from Bacillus thuringiensis subspecies dakota. SDS-PAGE demonstrates that Cry2Ae protein has a molecular weight of 71 kDa. Its sequence is modified for optimal expression in plants.
         
These events are combined by conventional breeding in another PIP, TwinLink(TM) that expresses Cry1Ab and Cry2Ae proteins to provide protection against feeding damage by lepidopteran insect larvae. 

The initial product characterization data submitted by BCS to support the FIFRA Section 3 Registration of PIP cotton Events T304-40, and GHB119 and their combination by conventional breeding in TwinLink(TM) were reviewed by the Agency and found Supplemental (U.S. EPA 02/04/11). There were several data deficiencies that the applicant had to address before EPA could continue its review for these events. Consequently, the data were considered insufficient to support the FIFRA section 3 registrations of cotton Events GHB119, T304-40, and TwinLink(TM) its respective combination PIP product (U.S. EPA DER dated February 4, 2011, from A. Waggoner to S. Bacchus and D. Greenway). Additional data, submitted by BCS in May 2011, were evaluated and found acceptable. 

For summaries of the data evaluated for product characterization for these PIPs, see Appendix A for Cry1Ab and Appendix B for Cry2Ae and TwinLink(TM) Cotton. The summaries include discussions of the following:  

   * Transformation System
   * Characterization of the DNA Inserted in the Plant and Inheritance and Stability
   * Protein Characterization 
   * Protein Equivalence 
   * Analytical Method
   * Product Characterization

As discussed in the summaries, Cry1Ab and Cry2Ae proteins are expressed at very low parts per million (ppm) levels in various parts of the transgenic cotton plant and are shown to be equivalent to the microbially expressed proteins. As terms of registration, EPA requires confirmation of the enforcement analytical methodology as discussed in Chapter VI of this BRAD. 

   B.  Human Health Assessment 

	1.	Toxicological Profile

EPA reviewed the available scientific data and other relevant information submitted in support of these actions to register Cry1Ab, Cry2Ae and TwinLink(TM) Cotton PIPs and considered their validity, completeness and reliability, and the relationship of this information to human risk. EPA has also considered available information concerning the variability of the sensitivities of major identifiable subgroups of consumers, including infants and children. 

All toxicology data requirements for the Bt cotton PIPs containing Cry 1Ab, Cry2Ae and their combined traits in TwinLink(TM) have been satisfied. Acceptable Tier I mammalian toxicology data/information support the proposed registrations of these active ingredients. Furthermore, Tier II and Tier III studies were not required for either Cry1Ab or Cry2Ae proteins, based on the lack of acute toxicity or pathogenicity in the Tier I studies. Refer to Tables A.3 and B.3 (in Appendices A, and B, respectively) for brief summaries of data evaluated to satisfy toxicity and allergenicity data requirements for these PIPs.


Acute toxicity

Cry1Ab and Cry2Ae proteins

The following summarizes the final determinations regarding acute toxicology data requirements for Cry1Ab and 2Ae proteins. See Appendices A and B in this BRAD for more detailed summaries. 


TABLE 1:  Tier 1 Mammalian Toxicology Data Requirements and Associated OCSPP Test Guidelines
                                     Study
                                  Conclusion
Acute Oral Toxicity/Pathogenicity 
       OCSPP 885.3050  -  acute oral LD50 >2000 mg/kg b.w. by tests.
Acute Dermal Toxicity
                          OCSPP 885.3100  -  Waived.
Acute Pulmonary Toxicity/Pathogenicity
                          OCSPP 885.3150  -  Waived.
Acute Injection Toxicity/Pathogenicity
OCSPP 885.3200  -  superseded by acute oral test for Cry2Ae protein (See Appendix B).
Hypersensitivity Incidents*
       OCSPP 885.3400  -  none reported yet but required if they occur.
Cell Culture
                 OCSPP 885.3500  -  only required for viruses.
Acute Oral Toxicity
          OCSPP 870.1100  -  waived for End-use Product TwinLink(TM).
Acute Dermal Toxicity
                            OCSPP 870.1200 - waived
Acute Inhalation Toxicity
                            OCSPP 870.1300- waived
Acute Eye Irritation
                            OCSPP 870.2400 - waived
Primary Dermal Irritation
                            OCSPP 870.2500 - waived

Hypersensitivity Incidents:  Reporting is required when incidents occur. No hypersensitivity incidents, including immediate-type or delayed-type reactions in humans or animals, occurred during the manufacture and field trials of the Bt Cry1Ab, Cry2Ae and their combination in TwinLink(TM) Cotton PIPs. Should any future hypersensitivity incidents occur, they must be reported to the Agency.

   a. Acute Oral Toxicity
Cry1Ab and Cry2Ae proteins

Acute oral toxicity Tier I studies in mice indicated that Cry 1Ab (U.S. EPA DER from A. Waggoner, 2011a to S. Bacchus and D. Greenway) and Cry2Ae (U.S.EPA DER from R. Edelstein, 2008 to S. Bacchus; A. Waggoner, 2011b to S. Bacchus and D. Greenway) proteins are non-toxic to laboratory mice. The acute oral toxicity levels of Cry1Ab and Cry2Ae proteins were assessed in separate tests. The mice were given purified, bacterially produced protein by oral gavage at a dosage of 2,000 milligrams/kilograms of body weight (mg/kg b.w.). All of the treated animals gained weight, and no signs of toxicity or other adverse effects occurred, and no test material-related findings were found at necropsy. The results of the study demonstrated that the acute oral LD50 of each of these proteins is greater than 2000 mg/kg b.w. 

The results of these studies demonstrate that no adverse effects are expected at a level that is well above the maximum possible exposure levels that are reasonably anticipated to be present in the crop (i.e. approximately 1 microgram/gram - equivalent to 1mg/kg - dry weight cottonseed  -  see Tables A.1 and B.1 in Appendices A and B). Cotton itself is not an edible food commodity in the human diet. Dietary exposure to processed cottonseed flour is also expected to be at lower levels than those for which no adverse effects were observed in these studies. 

When proteins are toxic, they are known to act via acute mechanisms and at very low dose levels (Sjoblad, Roy D. et al. (1992). Since no acute effects were shown to be caused by Cry1Ab or Cry 2Ae proteins at high dose levels in the acute oral toxicity tests, they are not considered to be toxic. Further, amino acid sequence comparisons showed no similarities between Cry1Ab or Cry2Ae proteins and known toxic proteins in protein databases that would raise a safety concern (U.S.EPA DERs from R. Edelstein, 2008 to S. Bacchus; from A. Waggoner, 2011a, 2011b to S. Bacchus and D. Greenway).
	
For microbial products, further Tiers II and III toxicity testing and residue data are only required if adverse effects are observed during Tier I tests. Based on the lack of acute oral toxicity and other adverse effects in the Tier I acute oral tests in mice, EPA did not require Tier II and Tier III testing for either Cry1Ab or Cry2Ae protein. This conclusion is similar to the Agency's position regarding toxicity testing and the requirement of residue data for the registered microbial Bt products containing the species from which these PIPs were derived (see 40 CFR §§ 158.2130(d)(1)(i) and 158.2140(d)(7)). 

   b. Allergenicity Assessment

Since Cry1Ab and Cry2Ae are proteins, allergenic sensitivities were considered for each in separate tests for each active ingredient. Currently, no definitive tests exist for determining the allergenic potential of novel proteins. Therefore, EPA uses a weight-of- evidence approach where the following factors are considered (U.S. EPA DERs 2011a and 2011b from Waggoner to Bacchus and Greenway; Appendices A and B): 

      1. Source of the trait. Bacillus thuringiensis, the microorganism from which the genetic materials encoding Cry1Ab and Cry2Ae proteins are derived, is not considered to be a source of allergenic proteins.
2. Amino acid sequence. Separate comparisons of the amino acid sequence of Cry1Ab and Cry2Ae proteins with known allergens showed no overall sequence similarity using the criterion of 35% identity over an 80 amino acid segment or identity at the level of eight contiguous amino acid residues. 
      3. Prevalence in food. Protein expression level analyses of Cry1Ab and Cry2Ae proteins are present at relatively low levels. Dietary exposure is extremely limited and expected to be correspondingly low. Cry1Ab and Cry2Ae expression in Event GHB119 leaf, root and pollen has been shown to be in the parts per million ranges, and cotton is not a directly consumed food commodity. 
      4. Digestibility. Cry1Ab protein was digested within 2 minutes, while the Cry2Ae protein was rapidly digested within 30 seconds in simulated mammalian gastric fluid (SGF) containing pepsin at a pH of 1.2. 
5. Glycosylation. Current scientific knowledge also suggests that common food allergens may be glycosylated [Codex Alimentarius Commission (CAC), 2003]. Neither Cry1Ab nor Cry2Ae protein expressed in cotton was shown to be glycosylated.

Considering all of the available information, EPA has concluded that the potential for Cry1Ab and Cry2Ae to be food allergens is minimal. 

   c. Tolerance Exemptions

Permanent tolerance exemptions have been established for residues of Cry1Ab and PAT, as discussed in Chapter II under REGULATORY BACKGROUND, above. 

EPA's reviews of the toxicology data support granting a permanent exemption from tolerance for residues of Cry2Ae protein in/on cotton food/feed commodities under section 408 of the Federal Food, Drug, and Cosmetic Act (FFDCA). On September 10, 2008, a temporary exemption from tolerance for residues of Cry2Ae PIP (40 CFR §174.530) was granted to BCS for cotton food/feed commodities in association with an Experimental Use Permit, EPA Reg. No. 264-EUP-143. The toxicological profile of Cry2Ae protein was previously described in the Federal Register (73 FR 52591; 9/10/2008: FRL-8380-1). Since then, BCS submitted additional data/information, and in 2009 petitioned EPA to establish a permanent exemption from the requirement of a tolerance for residues of Cry2Ae in or on all food commodities (see Chapter II, REGULATORY BACKGROUND). 

On the basis of evaluations of all relevant data submissions for Cry2Ae, EPA has determined that the data support amendment of 40 CFR §174.530 to establish a permanent tolerance exemption for residues of Cry2Ae in/on cotton food feed commodities. The exposure analysis and evaluation of additional data to establish this permanent exemption from tolerance are based upon the expression of Bacillus thuringiensis Cry2Ae protein in cotton. No other uses of this protein as a PIP in other food or animal feed commodities exist. As a result, there has been no effort to date to ensure that transformation events in plants other than cotton that express Cry2Ae protein have the same safety characteristics as those described in this evaluation. Consequently, the final tolerance exemption for Cry2Ae protein residues that the Agency is granting varies from what the petitioner sought in as much as it is limited to residues of Cry2Ae protein in/on certain cotton food/feed commodities when Cry2Ae protein is used as a PIP in cotton. In the future, additional uses in other food or animal feed crops may be proposed, and data or other information that demonstrate the same safety characteristics described for its expression in cotton must be submitted along with a petition to amend the tolerance exemption to include those uses.
 
2. Endocrine Disruptors

As required under FFDCA section 408(p), the Agency has developed the Endocrine Disruptor Screening Program (EDSP) to determine whether certain substances (including pesticide active and other ingredients) may have an effect in humans or wildlife similar to an effect produced by a "naturally occurring estrogen, or other such endocrine effects as the Administrator may designate." The EDSP employs a two-tiered approach to making the statutorily required determinations. Tier 1 consists of a battery of 11 screening assays to identify the potential of a chemical substance to interact with the estrogen, androgen, or thyroid (E, A, or T) hormonal systems. Chemicals that go through Tier 1 screening and are found to have the potential to interact with E, A, or T hormonal systems will proceed to the next stage of the EDSP where the Agency will determine which, if any, of the Tier 2 tests are necessary based on the available data. Tier 2 testing is designed to identify any adverse endocrine-related effects caused by the substance, and establish a quantitative relationship between the dose and the E, A, or T effect.

Between October 2009 and February 2010, the Agency issued test orders/data call-ins for the first group of 67 chemicals, which contains 58 pesticide active ingredients and 9 inert ingredients. This list of chemicals was selected based on the potential for human exposure through pathways such as food and water, residential activity, and certain post-application agricultural scenarios. This list should not be construed as a list of known or likely endocrine disruptors.

Cry1Ab and Cry2Ae proteins are not among the group of 58 pesticide active ingredients on the initial list to be screened under the EDSP. Under FFDCA section 408(p), the Agency must screen all pesticide chemicals. Accordingly, the Agency anticipates issuing future EDSP orders/data call-ins for all pesticide active ingredients. 

For further information on the status of the EDSP, the policies and procedures, the list of 67 chemicals, the test guidelines and the Tier 1 screening battery, please visit our website: http://www.epa.gov/endo/.

   1.    Food Quality Protection Act (FQPA) Considerations
   a. Aggregate Exposures 

In examining aggregate exposure, section 408 of FFDCA directs EPA to consider available information concerning exposures from the pesticide residue in food and all other non-occupational exposures, including drinking water from ground water or surface water and exposure through pesticide use in gardens, lawns, or buildings (residential and other indoor uses).

EPA considered available information on the aggregate exposure levels of consumers (including major identifiable subgroups of consumers) to the PIP residue and to other related substances. These considerations include dietary exposure under the tolerance exemption and all other tolerances or exemptions in effect for the PIP residue, and exposure from non-occupational sources. 

As discussed in Appendices A and B, the oral toxicity studies conducted at a dose of 2000 mg/kg b.w. in separate tests showed no adverse effects for either Cry1Ab or Cry2Ae protein, which were also shown to be rapidly digested in vitro. As previously stated, these proteins are expressed at very low ppm levels in the cotton PIPs, which are not directly consumed as food commodities. Although cotton is not a directly consumed food commodity, humans may be exposed to extremely low levels in the diet, potentially from ingestion of processed cotton products (e.g. cottonseed flour and oil). There is also a very remote possibility that Cry1Ab and Cry2Ae proteins can get in the water supply the same way that other proteins can migrate in crop debris from cotton into ground, and, maybe, drinking water. Such potential dietary exposure from cotton or drinking water is expected to be several orders of magnitude lower than the amounts of these proteins shown to have no toxicity in mammalian tests. EPA concludes that even negligible exposure via food and drinking water would present no harm, based on the lack of mammalian toxicity and allergenicity potential, and the rapid digestibility demonstrated for the PIPs. 
	
Non-occupational dermal and inhalation exposure is not expected, since the PIPs are expressed and contained within cotton plant cells. The uses of these PIPs are agricultural, so there would be no exposure to infants and children from residential, school or lawn use. The amino acid homology assessments for Cry1Ab and Cry2Ae proteins included similarity to known aeroallergens. It has been demonstrated that there is no evidence of occupationally related respiratory symptoms, based on a health survey on migrant workers after exposure to Bt pesticides (Bernstein et al. 1999). This observation is also relevant to the low potential for non-occupational inhalation exposure at levels far below those expected in occupationally exposed populations. 

Taking all these data and information into consideration, EPA concludes that even if negligible aggregate exposure should occur it would present no harm to the U.S. human population. 





   b. Cumulative Effects from Substances with a Common Mechanism of Toxicity

Section 408(b)(2)(D)(v) of FFDCA requires that, when considering whether to establish, modify, or revoke a tolerance, the Agency consider "available information" concerning the cumulative effects of a particular pesticide's residues and "other substances that have a common mechanism of toxicity."

Neither Cry1Ab nor Cry2Ae protein is considered toxic. They do not share a common mechanism of toxicity with any other substances, nor do they appear to produce a toxic metabolite produced by other substances. For the purposes of this assessment, therefore, EPA has assumed that Cry1Ab and Cry2Ae proteins do not have a common mechanism of toxicity with other substances. Thus, EPA concludes that there are no cumulative effects associated with Cry1Ab or Cry2Ae protein that need be considered. For information regarding EPA's efforts to determine which chemicals have a common mechanism of toxicity and to evaluate the cumulative effects of such chemicals, see EPA's website at http://www.epa.gov/pesticides/cumulative 
   
c.   Determination of Safety for U.S. Population, Infants and Children 
   
To evaluate human risk, EPA considered the validity, completeness, and reliability of the available data from the studies cited in Unit III regarding potential health effects for Cry2Ae protein. This evaluation included the low levels of expression of Cry2Ae proteins in cotton, as well as the lack of acute oral toxicity at high dose levels, heat stability, and in vitro digestibility of this protein. EPA also considered the minimal potential for allergenicity and the non-toxic source of the protein. Because of this lack of demonstrated mammalian toxicity, no protein residue chemistry data for Cry2Ae were required for a human health effects assessment. 
      
Finally, and specifically with regards to infants and children, FFDCA section 408(b)(2)(C) provides that EPA shall assess the available information about consumption patterns among infants and children, special susceptibility of infants and children to pesticide chemical residues, and the cumulative effects on infants and children of the residues and other substances with a common mechanism of toxicity. In addition, FFDCA section 408(b)(2)(C) provides that EPA shall apply an additional tenfold margin of safety for infants and children in the case of threshold effects to account for prenatal and postnatal toxicity and the completeness of the data base unless EPA determines that a different margin of safety will be safe for infants and children. 

Based on its review and consideration of all the available information, as discussed in this BRAD, EPA concluded that there are no threshold effects of concern and, as a result, that an additional margin of safety for infants and children is unnecessary in this instance.

   1. International Residue Limits
      
In making its tolerance decisions, EPA seeks to harmonize U.S. tolerances with international standards whenever possible, consistent with U.S. food safety standards and agricultural practices. In this context, EPA considers the international maximum residue limits (MRLs) established by the Codex Alimentarius Commission (Codex), as required by FFDCA section 408(b)(4). The Codex Alimentarius is a joint U.N. Food and Agriculture Organization/World Health Organization food standards program, and it is recognized as an international food safety standards-setting organization in trade agreements to which the United States is a party. EPA may establish a tolerance that is different from a Codex MRL; however, FFDCA section 408(b)(4) requires that EPA explain the reasons for departing from the Codex level. No Codex maximum residue level exists for the PIPs, Bacillus thuringiensis Cry1Ab or Cry2Ae protein or their combination Cry1Ab x Cry2Ae pyramided protein.

   A.  Environmental Effects Assessment 
      
Ecological Effects Data for Event T304-40 (Bt Cry 1Ab) and Event GHB119 (Bt Cry 2Ae)

All ecological and environmental Effects data requirements for the Bt cotton PIPs containing Cry1Ab, 2Ae proteins and their combined traits in TwinLink(TM) have been satisfied. For a more comprehensive discussion of the Agency's assessment of the data and information submitted concerning ecological and environmental risks for the registration of these PIPs (Cry1Ab, Cry2Ae, and their combination in TwinLink(TM) Cotton) refer to Appendix C. The tiered approach EPA uses to evaluate ecological and environmental effects, and summaries of the data evaluated, are included.
 	
In the absence of PIP-specific risk assessment guidance, EPA requires applicants for PIP registrations to meet the 40 CFR Part 158 data requirements for microbial pesticides. These requirements include testing on birds, mammals, nontarget insects, honey bee, plants, and aquatic species, and information has been submitted to address these requirements. Limit dose testing on representative organisms from several taxa was performed in support of the T304-40 and GHB119 Section 3 FIFRA registrations. As stated above, BPPD's risk assessments focus heavily on beneficial nontarget invertebrates, since they are most closely related to organisms susceptible to the insecticidal action of Bt toxins. The Cry1Ab and Cry2Ae proteins are meant to target species within the order Lepidoptera (moths and butterflies). 

Bt toxins are known typically to have a limited host range, however, to address any unforeseen change in activity spectrum as a result of laboratory protein synthesis and to fulfill the published registration data requirements EPA requires that test species used for non-target insect evaluations should include several invertebrate species that are not related to the target pests. Earthworm studies are also recommended. The toxicity of the Cry1Ab and Cry2Ae proteins has been evaluated on several species of invertebrates including the lady beetle, green lacewing, collembola, Daphnia, honey bee, and earthworm. Reproductive and/or developmental observations were also examined in the lady beetle, Daphnia, collembola, and honeybee studies. 

Since exposure may also occur to other nontarget organisms, EPA has received data to comply with the Agency's published non-target data requirements on other nontarget organisms. Tests to determine soil degradation of Cry1Ab and Cry2Ae proteins have also been submitted. The individual results for nontarget organism and soil degradation testing for Cry1Ab and Cry2Ae proteins, and where applicable, for their combination Cry1Ab x Cry2Ae in TwinLink(TM) Cotton, are summarized in Appendix C. Full reviews of each study for each event can be found in the individual Data Evaluation Records. 

The October 2000 SAP recommended that while actual plant material is the preferred test material, bacteria-derived protein is also a valid test substance, particularly in scenarios where test animals do not normally consume cotton plant tissue and where large amounts of Cry protein are needed for maximum hazard dose testing. In support of the T304-40 and GHB119 registrations, test substances used in the submitted studies included bacteria-produced purified Cry1Ab and Cry2Ae protein, respectively. Comparative analyses have been performed to verify the equivalence of the bacterially-produced and purified Cry1Ab and Cry2Ae and the Cry1Ab and Cry2Ae proteins produced in T304-40 or GHB119, respectively, and TwinLink(TM) cotton plants. Product characterization data demonstrate the bioequivalence of the plant-expressed and microbially expressed proteins (Waggoner 2011a, 2011b). Consequently, on the basis of the environmental risk assessment and evidence of bioequivalence, EPA concludes that Cry1Ab and Cry2Ae PIPs are not likely to pose a hazard to nontarget organisms.

On the basis of the ecological effects assessment, EPA concludes that significant adverse effects are not expected on birds, mammals, nontarget insects, honey bees, freshwater and marine/estuarine fish and invertebrates, and terrestrial and aquatic plants as a result of the use of Event T304-40 and Event GHB119 in cotton. EPA has also determined that the potential for gene capture and expression of Cry1Ab or Cry2Ae protein by wild or weedy relatives of cotton within the U.S. or its territories does not pose a significant risk as long as the current sales and distribution restrictions on Bt cotton remain in place. Available data indicate that Cry proteins do not have any measurable adverse effect on microbial populations in the soil, nor has horizontal transfer of genes from transgenic plants to soil bacteria been demonstrated.

Endangered Species

Because of the selectivity of Cry1Ab and Cry2Ae proteins for lepidopteran species, endangered species concerns are also expressed for insect species in the order Lepidoptera. The Agency has previously determined that any potential concern regarding range overlap of endangered Lepidoptera with cotton production was mainly restricted to the Kern primrose sphinx moth (Euproserpinus euterpe). However, the Agency determined that cotton is not a host plant for this species nor do host-range considerations for this species or other endangered insects place habitat in or near cotton fields (US EPA 2008a).     

As previously noted, there is a possibility for gene transfer in locations where wild or feral cotton relatives exist. As a result, EPA requires stringent sales and distribution restrictions on Bt cotton within these areas to preclude outcrossing or hybridization from the crop to sexually compatible relatives. Therefore, EPA does not expect that any threatened or endangered species will be affected by outcrossing to wild relatives or by competition with such entities.

Cry1Ab protein has not been shown to have toxic effects on mammals, birds, freshwater invertebrates, terrestrial nontarget invertebrates, and plants, and exposure is not anticipated in aquatic environments. EPA, therefore, makes a "No Effect" determination for direct and indirect effects to listed species of these taxa for Cry1Ab protein expressed in Event T304-40 cotton. 

Since Cry2Ae protein has not been shown to have toxic effects on birds, mammals, freshwater invertebrates, terrestrial nontarget invertebrates, and plants, and also because exposure is not anticipated in aquatic environments, the Agency makes a "No Effect" determination for direct and indirect effects to listed species of these taxa for Cry2Ae protein expressed in Event GHB119 cotton. 
                  
   D.   Insect Resistance Management

The goal of insect resistance management (IRM) is to mitigate the potential risk of resistance in Bt crops, including Bt cotton. IRM typically involves two phases:  1) data submitted to assess the risk of resistance for a certain Bt toxin and crop including information and data on pest biology, dose expression of the toxins in the PIP, simulation modeling, cross resistance potential and baseline susceptibility of the target pests to the toxin(s); and 2) a post-registration stewardship program to implement the resistance mitigation measures which includes resistance monitoring, remedial action (should resistance occur), grower compliance with IRM requirements, and grower education. The standard risk mitigation strategy for IRM has been the use of structured refuges, which are non-PIP portions of the crop (or other suitable plant hosts) that provide a source of susceptible insects to mate with any resistant insects emerging from the PIP field (thus diluting the frequency of resistance genes in the pest population). Some Bt cotton products have been approved with "natural refuge" instead of structured refuges. Natural refuges include non-cotton plant hosts of the target pests such as weeds, wild hosts, or other cultivated crops (e.g., soybean, peanut, tobacco). 

BCS submitted an IRM plan and data with their application for registration of TwinLink(TM) cotton (MRID#s 47640014 and 47634823). TwinLink(TM) is a pyramided Bt cotton PIP expressing Cry2Ae and Cry1Ab proteins and is targeted against three major cotton pests: tobacco budworm (TBW, Heliothis virescens), cotton bollworm (CBW, Helicoverpa zea), and pink bollworm (PBW, Pectinophora gossypiella). BPPD reviewed this initial submission and identified a number of deficiencies (review dated April 22, 2010). BCS responded to these deficiencies with several resubmissions (MRID#s 48480010, 48480011, and 484951-01) that were reviewed by BPPD (reviews dated September 28, 2010 and June 28, 2011). BPPD concluded that these resubmissions satisfied the outstanding deficiencies and that the IRM proposal for TwinLink(TM) cotton was acceptable.

BPPD's major IRM conclusions from its assessment of TwinLink(TM) Cotton are detailed below. The IRM assessment included data developed for TwinLink(TM) Cotton (Cry2Ae x Cry1Ab) as well as its single toxin constituent PIPs, GHB119 (Cry2Ae) and T304-40 (Cry1Ab).

 1. DOSE

   a. BPPD concludes that the diet incorporated and overlay assays together (using verification method 1) suggest that TwinLink(TM) Cotton (Cry2Ae x Cry1Ab proteins) and GHB119 (Cry2Ae protein) may express a high-dose against CBW but that T304-40 (Cry1Ab protein) may express less than high-dose against the target pest.
      
   b. BPPD concludes that diet incorporated and overlay assays (using verification method 1) together suggest that Events GHB119 (Cry2Ae protein) and T304-40 (Cry1Ab protein) may express a high-dose and TwinLink(TM) (Cry2Ae x Cry1Ab proteins) may express an effective high-dose against TBW. 
      
   c. BPPD concludes that the single toxin Events GHB119 (Cry2Ae) and T304-40 (Cry1Ab) both express less than a high dose against PBW, but that TwinLink(TM) (Cry2Ae x Cry1Ab) expresses a high dose against the insect.

2. CROSS RESISTANCE

   a. Based on submitted data, BPPD concludes that there is low cross resistance potential between the Cry2Ae and Cry1Ab toxins that are expressed in TwinLink(TM).
       
   b. BPPD notes that there is a potential for cross resistance between Cry2Ae and Cry2A2 proteins (registered in Bollgard II cotton). Should resistance develop to either toxin, the second toxin could also be compromised.

   c. Similarly, there is cross resistance potential between Cry1Ab (also registered in VipCot cotton) and Cry1Ac (registered in Bollgard and WideStrike cotton) and Cry1F (registered in WideStrike cotton). Competitive binding assays have shown that these toxins share binding sites in insect midgets which can lead to cross resistance (though the potential is somewhat less for Cry1F).


3. SIMULATION MODELING

   a. BCS conducted separate models to evaluate the durability of TwinLink(TM) for CBW and TBW. Both models were intended to evaluate durability with "natural refuges" (i.e., non-cultivated plant hosts such as weeds or cultivated crops other than cotton that can serve as sources of susceptible insects).
      
   b. Modeling for CBW was conducted by Dr. Michael Caprio (Mississippi State University). The model was deterministic with a probabilistic function in which parameters were established using PERT distributions (i.e. a distribution in which the mean value is weighted more heavily than minimum or maximum values). BPPD had a number of concerns regarding some of the modeling assumptions including dose mortality, initial resistance allele frequency, and the lack of inclusion of Cry1Ab corn in the model landscape. BCS responded to these deficiencies with revised modeling that employed more conservative parameter values. The revised model was run for 1,000 simulations; resistance did not evolve to either Cry1Ab or Cry2Ae proteins (within the 30 year time horizon of the model) in any of the simulations.

   c. BCS conducted two modeling analyses for TBW:  i) a stochastic, spatially-explicit model, and ii) a deterministic, spatially-implicit model. As with CBW, the TBW models were designed and conducted by Dr. Michael Caprio. The modeling generally took a conservative approach by including "worst case" simulations for certain parameters (dose mortality and dominance) and incorporated other Cry1A cotton products in the landscape. Results from the stochastic model showed that resistance did not evolve during the 31 year model time frame. However, when dominance values (for resistance alleles) and late season survival were increased, the time to resistance decreased. With the deterministic model, the time to resistance was 224 years (both toxins) under the benchmark parameters. Deterministic simulations with worst case assumptions included resulted in a durability estimate of 80 years (a 2.8 fold difference from the benchmark simulations). A separate deterministic simulation was also run with a single trait Bt cotton product (Cry1Ac) with a 20% structured refuge that resulted in an estimated durability of 18.8 years.
      
   d. Based on the modeling results, BPPD concluded that TwinLink(TM) cotton deployed with a natural refuge should not be expected to pose a greater risk of resistance evolution than other Bt cotton products (e.g., Bollgard II) currently registered with a natural refuge strategy.
   
   
4. REFUGE STRATEGY
   
   a. BCS provided adequate information to demonstrate that the use of a natural refuge with TwinLink(TM) cotton should provide high durability for both toxins (Cry1Ab and Cry2Ae). The risk of resistance for TwinLink(TM) is not expected to be greater than other registered Bt cotton products (Bollgard II, WideStrike, VipCot) that employ a natural refuge strategy.
      
   b. The use of natural refuge has been assessed only for southeastern cotton growing regions. Western cotton regions (including west Texas, New Mexico, Arizona, and California) should still employ structured refuges (as have been required for other Bt cotton products). The structured refuge options are :  i) 5% external, unsprayed non-Bt cotton refuge, ii) 20% external, sprayed non-Bt cotton refuge, iii) embedded refuge  -  either 5% or 1 row of non-Bt cotton per 6-10 rows of Bt cotton.

   c. Registrations have been proposed for Events GHB119 (Cry2Ae) and T304-40 (Cry1Ab), both of which are single trait products. These products are "breeding" registrations (for the production of TwinLink(TM)) and are not intended to be commercially sold. BCS has requested that no refuge be required for these products. Breeding registrations typically do not have refuge requirements, however, BPPD recommends the use of acreage/geographic limitations for the single trait products as a means to mitigate potential resistance development. 
      

5. RESISTANCE MONITORING AND REMEDIAL ACTION

   a. BCS has indicated that they are developing baseline susceptibility data for TBW, CBW, and PBW to Cry1Ab and Cry2Ae proteins. They plan to have the studies completed in 2012 (i.e., after product registration). Given the importance of baseline data to the resistance monitoring program, BPPD recommends that these data be generated by the first growing season of commercial use of TwinLink(TM). 
      
   b. BCS has proposed to monitor for resistance with all three target pests in areas of high TwinLink(TM) adoption. BPPD agrees with this proposal but also recommends that monitoring efforts also focus on regions in states with little natural refuge (e.g. Mississippi). 

   c. For PBW, BCS plans to use the remedial action plan developed by the Arizona Bt Cotton Working Group (which has been employed for other registered Bt cotton products). For TBW and CBW, the remedial action plan will be refined once baseline susceptibility data are available (i.e., to define "suspected" and "confirmed" resistance). The base elements of the plan proposed by BCS are similar to other Bt cotton products; however, BPPD recommends that the plan be adopted to include a provision that EPA will be informed within 30 days of a confirmed resistance event."  Additionally, a stipulation should be added that, in the event of resistance, BCS will develop a long-term resistance management action plan according to the characteristics of the resistance event(s) and local agronomic practices.

   d. BPPD recommends that BCS submit final versions of the resistance monitoring and remedial action plans for all pests as a term or condition of registration.


6. COMPLIANCE

   a. A natural refuge strategy for TBW and CBW would eliminate the need for growers to plant a refuge where alternate hosts are abundant. However, there are cotton-growing regions where natural refuge is not applicable and where planting a refuge is still required (i.e., western states). For those regions it is recommended that BCS develop, implement, and report to EPA on programs to evaluate and promote growers' compliance with IRM requirements. 
        
   b. Grower compliance in regions where structured refuge is required should be assessed with on-farm assessments in addition to anonymous phone (or internet) surveys. BPPD recommends that BCS use the existing grower compliance paradigm that has been established for other Bt cotton registrations.
      
7. GROWER EDUCATION

      a.       It is recommended that BCS develop a grower education program that targets states and counties where a non-Bt corn variety must be planted as the refuge for TwinLink(TM) cotton (i.e., structured refuge). 

      b.       It is also recommended that BCS develop a grower guide that at a minimum describes the technology (including the concept of natural refuge) and includes unexpected damage guidelines.  
         
      
8. ANNUAL REPORTING

BPPD recommends that BCS submit annual reports for the following:
         i) Sales data;
         ii) Annual resistance monitoring results; and 
         iii) Compliance results for cotton growing regions where Natural Refuge is not applicable.

Overall, BPPD concluded that BCS has submitted adequate resistance management data and information to support commercial use of the product with a natural refuge. The risk of resistance for TwinLink(TM) should be no greater than for other currently registered Bt cotton pyramids (including Bollgard II, WideStrike, and VipCot). There are no data gaps, though it is recommended that BCS submit additional information on resistance monitoring (baseline data and a final monitoring plan), remedial action, compliance, and grower education (i.e., grower guides) as described above.   

A more detailed discussion of IRM including full summaries of BPPD's reviews of the supporting data for Cry1Ab, Cry2Ae and TwinLink(TM) Cotton PIPs are available as Appendix D of this BRAD.
IV.	ENVIRONMENTAL JUSTICE

EPA seeks to achieve environmental justice, the fair treatment and meaningful involvement of all people, regardless of race, color, national origin, or income, in the development, implementation, and enforcement of environmental laws, regulations, and policies. To help address potential environmental justice issues, the Agency seeks information on any groups or segments of the population who, as a result of their location, cultural practices, or other factors, may have atypical, unusually high exposure to these PIP proteins Cry1Ab, Cry2Ae and their combination in TwinLink(TM) compared to the general population. No comments were received during the public comment period.

For additional information regarding environmental justice issues, please visit EPA's web site at http://www.epa.gov/compliance/environmentaljustice/index.html.
V. 	RISK MANAGEMENT AND PROPOSED REGISTRATION DECISION

Section 3(c)(5) of FIFRA provides for the registration of a new active ingredient if it is determined that (A) its composition is such as to warrant the proposed claims for it; (B) its labeling and other materials required to be submitted comply with the requirements of FIFRA; (C) it will perform its intended function without unreasonable adverse effects on the environment; and (D) when used in accordance with widespread and commonly recognized practice, it will not generally cause unreasonable adverse effects on the environment. 

The four criteria of the Eligibility Determination for Pesticidal Active Ingredients are satisfied by the science assessments supporting the PIPs containing Bacillus thuringiensis Cry1Ab, Cry2Ae and their combination, Cry1Ab x Cry2Ae, in TwinLink(TM). These PIPs are not expected to cause unreasonable adverse effects and are likely to provide protection as claimed when used according to label instructions. The Agency believes, therefore, that Bacillus thuringiensis Cry1Ab, Cry2Ae and their combination, Cry1Ab x Cry2Ae, in TwinLink(TM) are eligible for registration for the proposed uses. 
VI.   TERMS AND CONDITIONS OF THE REGISTRATIONS

As terms of registration, the applicant must submit or provide the following:
 
      A. Analytical method: OCSPP 860.1340 (40 CFR § 158.2120)  

The following validation studies must be conducted by an independent third party laboratory, as a term of the proposed registrations: 

   1. Term of Registration for Cry1Ab - Event T304-40 (EPA File Symbol 264-RNOU) 
      Verification of the specific ELISA test kit to be used as the proposed analytical method for the detection of Cry1Ab protein residues expressed in Event T304-40 cotton is needed. In addition, a validation study conducted by an independent third party laboratory is also needed to evaluate the performance of the designated analytical method for the detection of Cry1Ab protein residues expressed in cotton Event T304-40. This study should use the prescribed methodology and the reagents necessary for validating the analytical method used in the test kit and should be conducted according to the study parameters in the EPA residue chemistry testing guideline OCSPP 860.1340 Residue Analytical Method and Pesticide Registration (PR) Notice 96-1: Tolerance Enforcement Methods - Independent Laboratory Validation by Petitioner (February 7, 1996). (Optional: Consultation with the Agency is highly recommended)

   2. Term of Registration for Cry2Ae- Event GHB119 (EPA File Symbol. 264-RNOL) 
      A validation study conducted by an independent third party laboratory is still needed to evaluate the performance of the ELISA test kit as the designated analytical method for the detection of Cry2Ae protein residues expressed in Event GHB119 cotton. This study should use the prescribed methodology and the reagents necessary for validating the analytical method used in the test kit and should be conducted according to the study parameters in the EPA residue chemistry testing guideline OCSPP 860.1340 Residue Analytical Method and Pesticide Registration (PR) Notice 96-1: Tolerance Enforcement Methods - Independent Laboratory Validation by Petitioner (February 7, 1996). (Optional: Consultation with the Agency is highly recommended)

   3. Term of Registration for TwinLink(TM) (T304-40 x GHB119) combination PIP product (EPA File Symbol 264-RNOA). Any outstanding data submitted to satisfy the terms of registration for PIP cotton Events T304-40 and GHB119 (EPA File Symbol 264-RNOU and 264-RNOL, respectively) must be submitted and found acceptable in order to complete the database for TwinLink(TM) cotton, as the associated combination PIP product developed by conventional breeding of the parental Events T304-40 x GHB119 cotton

      A. Insect Resistance Management

   1. BCS must submit a detailed resistance monitoring plan for the major pests of TwinLink(TM)  cotton:  tobacco budworm, cotton bollworm, and pink bollworm. In addition, baseline susceptibility and diagnostic concentration determinations for tobacco budworm, cotton bollworm, and pink bollworm to Cry2Ae and Cry1Ab must be submitted. 

   2. BCS must submit a final remedial action plan for tobacco budworm and cotton bollworm. The remedial action plan must include definitions of "suspected" and "confirmed" resistance and steps to take in the event of confirmed resistance.

   3. BCS must submit a compliance assurance program (CAP) for TwinLink(TM) that must include a "phased compliance approach" that outlines instances of non-compliance to the IRM requirements and options of responding to non-compliant growers. The CAP must be developed for regions that have structured refuge requirements (i.e., western cotton-growing regions); a CAP is not necessary for regions with natural refuge requirements.

   4. Persons purchasing TwinLink(TM) cotton must sign a grower agreement. The term "grower agreement" refers to any grower purchase contract, license agreement, or similar legal document and must clearly set forth the terms of the current IRM program. By signing the grower agreement, a grower must be contractually bound to comply with the requirements of the IRM program. In addition, BCS must implement a system which is reasonably likely to assure that persons purchasing TwinLink(TM) will affirm annually that they are contractually bound to comply with the requirements of the IRM program. BCS must submit a copy of the grower agreement/stewardship documents and a written description of a system assuring that growers will sign the grower agreement.

      A. Adverse Effects

Reports of all incidents of adverse effects to the environment must be submitted to the Agency under the provisions stated in FIFRA Section 6(a)(2).

      B. Hypersensitivity Incidents

All incidents of hypersensitivity (including both suspected and confirmed incidents) must be reported to the Agency under the provisions of 40 CFR §158.2140(d).

      C. Annual Reporting

As discussed under IRM in Chapter IV of this BRAD, BPPD recommends that BCS submit annual reports for the following:
   a. Sales data;
   b. Annual resistance monitoring results; and 
   c. Compliance results for cotton growing regions where Natural Refuge is not applicable.APPENDIX A:  Product characterization and human health effects  -  Bacillus thuringiensis Cry1Ab (Event T304-40) in cotton plant incorporated protectant.

A.1. BACKGROUND

BCS has developed a cotton line from Event T304-40 (EPA File Symbol 264-RNOU and OECD Unique Identifier: BCS-GHØØ4-7) that express a Cry1Ab plant-incorporated protectant (PIP) protein derived from Bacillus thuringiensis (Bt) subspecies berliner, which is intended to control lepidopteran larvae such as bollworm (CBW, Helicoverpa zea), tobacco budworm (TBW, Heliothis virenscens) larvae, and fall armyworm (FAW, Spodoptera frugiperda). BCS previously submitted an EUP application to the Agency for conducting experimental field trials on Event T304-40 cotton, which was granted on February 7, 2006. It was extended on March 8, 2007 to expire on December 31, 2011.  

The product characterization and human health data submitted for Event T304-40 PIP cotton were reviewed and found acceptable for the EUP applications (US EPA, 2006). BCS later submitted a FIFRA section 3 registration application containing new product characterization data and updated toxicological data for Cry1Ab protein. These data were reviewed by the Agency and several product characterization studies were found supplemental. BCS submitted additional data to address the data deficiencies identified in those studies.

Summaries of the resubmission data for product characterization, as well as the remaining toxicological data submitted to register Event T304-40, are provided in this Appendix. A number of studies were reviewed under a work-share program, under an EPA-BPPD and USDA-BRS Memorandum of Understanding (09-2000-0052-MU). For more detailed information, please refer to the individually reviewed studies, summarized in individual Data Evaluation Reports (DERs), by the assigned MRID numbers (US EPA, 2011). 

Event T304-40 cotton expresses Cry1Ab protein as an active ingredient and phosphinothricin acetyltransferase (PAT) enzyme as an inert ingredient. The status of the existing tolerance exemptions for Cry1Ab protein (40 CFR 174.511) and PAT (40 CFR 174.522) has been previously discussed in Chapter IIA.2.b. of this BRAD. Residues of Bt Cry1Ab protein are exempt from the requirement of a tolerance when used as plant-incorporated protectants in all food commodities and the PAT enzyme is exempt from the requirement of a tolerance when used as plant-incorporated protectant inert ingredient in all food commodities. The data and information discussed below support these existing tolerance exemptions.

A.2. PRODUCT CHARACTERIZATION [40 CFR § 158.2120(c)]

1) Event T304-40 Cotton expressing Cry1Ab protein

a. Transformation System

Event T304-40 was created through Agrobacterium-mediated transformation of cotton variety Coker 315 with plasmid pTDL008. The cry1Ab gene sequence was modified for expression in plants and consists of one nearly complete copy of the T-DNA flanked by an inverted incomplete copy of the cry1Ab gene cassette and one additional 3'me1 terminator. The Cry1Ab protein encoded by the gene inserted into cotton is expected to have 617 amino acids and a molecular weight of approximately 69 kDa. In addition, the deduced amino acid sequence of BCS's Cry1Ab protein is identical to the amino acid sequence of the Cry1Ab protein in a previously registered PIP (Event Bt11 corn, OPP chemical code 006444) with the exception of one amino acid at the C- terminus and two amino acids at the N-terminus (MRID No. 467474406; US EPA 2006). Plasmid pTDL008 also contains the bar gene as a selectable marker, which is derived from the soil microorganism, Streptomyces hygroscopicus (described by Thomson et al. 1987). The bar gene encodes the enzyme phosphinothricin acetyltransferase (PAT), which is co-expressed with Cry1Ab protein in Event T304-40 cotton as a PIP inert ingredient. The PAT enzyme provides the plant with tolerance to herbicide applications containing glufosinate-ammonium by acetylating the herbicide glufosinate-ammonium and making it unable to bind glutamine synthase. The PAT enzyme encoded by the bar gene is a homodimer of 183 amino acids and has an expected molecular weight of approximately 22 kDa. 

b. Characterization of the DNA Inserted in the Plant and Inheritance and Stability

Characterization of the DNA isolated from Event T304-40 cotton using restriction enzyme digests and Southern blot analyses as well as DNA sequencing indicates that the transgenic insert was inserted in the cotton genome, consisting of one nearly complete copy of the T-DNA flanked by an inverted incomplete copy of the cry1Ab gene and one additional 3'me1 terminator. No vector backbone sequences from the transformation plasmid pTDL0008 were found in T304-40. Inheritance and stability studies of the cry1Ab gene in Event T304-40 verified that it is stably integrated into the cotton genome, segregating in an expected Mendelian fashion.

c. Protein Characterization 

Event T304-40 cotton expresses Cry1Ab protein produced by Bacillus thuringiensis subsp. berliner. The Cry1Ab protein after cleavage of the transit peptide has 617 amino acids. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analyses of Cry1Ab protein, as expressed in Event T304-40 showed a molecular weight close to the deduced MW for Cry1Ab protein and Western blot analyses showed a single immunoreactive band. Cry1Ab protein was also found to be non-glycosylated. N-terminal sequencing also confirmed that the transit peptide of the plant form was cleaved at the expected residue. The biological activity of Cry1Ab protein was assessed by an insect feeding bioassay using Heliothis virescens first instar larvae and the LC50 for the plant- produced Cry1Ab protein was 74 ng/cm². 

d. Protein Expression

Expression level data were provided for Cry1Ab protein in different plant tissues and at different growth stages in Event T304-40 cotton and a summary of the results are provided in Table A.1 below. The data were produced using an ELISA method. Grain is the same as delinted seed, also commonly referred to as cottonseed. The distinction is made when distinguishing between grain intended for processing into cottonseed etc. or seed intended to be planted to grow cotton plants.


   Table A.1. Mean Cry1Ab Protein Expression levels in Event T304-40 Cotton
                                  Tissue Type
                             A.1. Cry1Ab Protein 
                           (ug/g dry weight +- SD)*
Roots
                          3.8  +- 0.9  -   8.6 +- 2.6
Stems
                         0.88 +- 0.37  -   6.2 +- 1.1
Leaves
                         0.45 +- 0.08   -   4.6 +- 0.6
Squares
                                  2.90 +- 1.0
Apex
                                  1.3 +- 0.3
Bolls
                                  0.38 +- 0.2
Whole plants
                                  2.2 +- 1.1
Pollen 
                                     NA[a]
Nectar 
                                     NA[a]
Flowers
                                  10.6 +- 1.9
Cottonseed
                                  4.10 +- 1.6
   		*Range reflects means at different growth stages           
   		a Not applicable due to the insufficient availability of material.
   
e. Analytical Method

BCS CropScience has provided validation studies for commercial enzyme-linked immunosorbent assay (ELISA) test kits for the analytical detection of Cry1Ab protein residues in cottonseed and cotton leaf tissue. Results showed the test kits are able to detect Cry1Ab protein residues in cotton with sufficient accuracy, precision, and sensitivity. However, verification of the specific ELISA test kit method to be used as the proposed residue analytical method for the detection of Cry1Ab protein in Event T304-40 cotton is needed. In addition, a validation study conducted by an independent 3[rd] party laboratory is needed to evaluate the performance of the designated residue analytical method for the detection of Cry1Ab protein expressed in cotton Event T304-40. 

Table A.2 summarizes the product characterization EPA evaluated and found acceptable for the FIFRA section 3 registration of Cry1Ab protein. 

Table A.2. Product Characterization Data for Event T304-40 cotton expressing Bt Cry1Ab protein
                          (EPA File Symbol 264-RNOU)
                                  Study Title
           A.2. Product Characterization - Bt Cry1Ab Protein Summary
                                    MRID No
Characteristics of Cry1Ab cotton plants derived from transformation Events number T303-3 and T304-40
This submission replaces MRID 46474401 and includes descriptions of the genetic constructs used in Events T303-3 and T304-40, the sources of the introduced genes, the modes of action of the expressed proteins, and preliminary molecular analysis, including the protocol used and Southern blot data. The molecular analysis indicated that in each event, one copy of the cry1Ab gene was inserted into cotton (Gossypium hirsutum). 

CLASSIFICATION:  SUPPLEMENTAL. Data gaps addressed in MRID No. 48480003.
                                   46708801
                   [This study supersedes MRID No. 46474401]
Characteristics of Cry1Ab cotton plants derived from transformation Events number T303-3 and T304-40[1]
This submission includes descriptions of the genetic constructs used in Events T303-3 and T304-40, the sources of the introduced genes, the modes of action of the expressed proteins, and preliminary molecular analysis. The study authors state that the molecular analysis indicated that in each event, one copy of the cry1Ab gene was inserted into cotton (Gossypium hirsutum).

CLASSIFICATION:  SUPPLEMENTAL  -  Data gaps addressed in MRID No. 48480003.
                                   46474401
PAT and Cry1Ab protein in cotton tissues of T303-03 and T304-40 Events, preliminary report[1]
PAT and Cry1Ab protein in cotton tissues of T303-03 and T304-40 Events, preliminary report -- BCS conducted preliminary protein expression analysis in support of the EUP request for Cry1Ab cotton Events T303-.3 and T304-40. Average  Cry1Ab protein and PAT enzyme concentrations, as determined by ELISA in ground cotton seed from Events T303-3 and T304-40, were 16 g/g and 2.2 g/g fresh weight, respectively, for  Cry1Ab  protein and 117 and 129 g, respectively, for PAT. The dry weight equivalents in seeds are: 2 and 16 g  of  Cry1Ab protein per 0.91g dry seed (or 0.0002 and 0.0017% dry weight) and 117 and 129 g  of PAT per 0.91g dry seed (or 0.013 and 0.014% dry weight). Cry1Ab protein concentrations in early flowering cotton leaf tissues grown in a greenhouse averaged 1.9 g/g fresh weight for T303-3 and 0.3 g/g for T304-40. In leaf tissue collected six weeks after flowering, the Cry1Ab protein concentrations were 2.1 g/g fresh weight for Event T303-3 and 0.16 g/g for Event T304-40. The dry weight content of Cry1Ab protein in leaves was estimated to be: 0.2 to 2.3 g  of  Cry1Ab  protein per 0.12 g dry leaf. From Event T303-3 and T304-40 cotton grown in the field, Cry1Ab protein was detected in the leaves, squares, flower and boils at concentrations ranging from 0.01 to 0.09 % total soluble protein. The PAT content of cotton tissues (other than seed) was not presented in the study report.

CLASSIFICATION:  SUPPLEMENTAL- No additional data needed.
                                   46708804 
                                       
                   [This study supersedes MRID No. 46474404]
Product Characterization of Cry1Ab Cotton Event T304-40

The purpose of this study was to determine if the transfer DNA (T-DNA) containing the intended cry1Ab gene from plasmid pTDL008 was inserted into the genomic DNA of cotton Event T304-40, and the backbone DNA from pTDL008 was not inserted into cotton Event T304-40. Southern blot analyses showed that the inserted transgenic sequence in T304-40 consisted of one nearly complete copy of cry1Ab flanked by an inverted, incomplete copy of the cry1Ab gene cassette, and one additional 3' me1 terminator. Southern blot analysis for the presence of vector backbone in T304-40 was negative using 4 of the 7 backbone probes. 

However, there were several data missing from this report and several regions of the vector backbone sequence were not confirmed as absent after the transformation process to create T304-40 cotton. Additional analyses such as demonstrating the cry1Ab gene was stably integrated and segregated according to Mendelian laws of inheritance over several successive generations of Event  T304-40  cotton were not addressed in this study.

CLASSIFICATION:  SUPPLEMENTAL- Data gaps addressed in MRID No. 48480005.
                                  47634801  
Structural and Functional Equivalence of Cry1Ab and PAT/bar Proteins Produced in Escherichia coli to Cry1Ab and PAT/bar Proteins from Event T304-40 and TwinLink(TM) Cotton, Gossypium hirsutum[2]
The purpose of this study was to demonstrate the biochemical and functional equivalency of the Cry1Ab and PAT/bar proteins from cotton Event T304-40 with the Cry1Ab and PAT/bar proteins expressed in Escherichia coli and TwinLink(TM) cotton, based on molecular weight and immunoreactivity. SDS-PAGE analyses of Cry1Ab protein isolated from Event T304-40 cotton leaves showed a very faint band estimated to be at a similar molecular weight as the band exhibited for the E. coli-derived Cry1Ab protein standard (~66 kDa). Western blot analyses showed immunologically reactive bands at the same position (~66kDa) for Cry1Ab protein expressed in cotton Event T304-40 and TwinLink(TM) to the E. coli-derived Cry1Ab protein. However, it is unclear in the study report if the biochemical data presented in the study were derived from the same source of the bacterial-expressed Cry1Ab protein test substance. SDS-PAGE analyses showed identical electrophoretic mobilities and similar molecular weights (~20 kDa) for the PAT/bar protein isolated from Event T304-40  cotton leaves, from the PAT/bar E. coli standard and the TwinLink(TM) cotton. Western blot analyses showed immunologically reactive bands at the same position (~20 kDa) for all three PAT protein sources. Several single-dose sensitive insect bioassays were conducted to confirm bioactivity, but these data did not provide the quantitative results to allow an adequate comparison of the insecticidal activity for each source of protein.
In general, the study report lacked sufficient details on the materials/ methods used, the experimental design, and reported results to properly document the bioequivalence of the bacterial-derived Cry1Ab protein test substance to the T304-40 cotton plant-expressed Cry1Ab protein. Therefore, the biochemical and functional equivalence of the plant- and bacterial- expressed protein sources cannot be confirmed from these data.   

CLASSIFICATION:  SUPPLEMENTAL- Data gaps addressed in MRID No. 48480004.
                                  47634805  
Analyses of Raw Agricultural Commodity (Fuzzy Seed) of Cry1Ab Cotton Event T304-40 for PAT/bar and Cry1Ab and its Non-transgenic Counterpart for PAT/bar and Cry1Ab Proteins[2]
The purpose of this study was to quantify the expression of Cry1Ab and PAT proteins in the fuzzy seeds of transgenic cotton Event T304-40 and in a non-transgenic cotton line, Coker 312. Fuzzy seed was grinded into fractions of lint coats and kernels and analyzed separately for Cry1Ab and PAT proteins and total extractable protein. To quantitate the amount of the protein analytes in the fuzzy seeds, ELISA results of separate assays were combined for the kernel and lint coat fractions. 

The average Cry1Ab protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, was 0.955 +-0.317 ug/g and 1.03 +- 0.378 ug/g, respectively. As a percent of total crude protein, Cry1Ab protein comprised an average of 0.000434 +- 0.000167% and 0.000465 +- 0.000191% of the total crude protein in unsprayed and sprayed fuzzy seed, respectively. The average PAT protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, was 95.4 +- 17.0 ug/g and 100 +- 16.8  ug/g, respectively. As a percent of total crude protein, PAT protein comprised an average of 0.0428 +- 0.00810% and 0.0445 +- 0.00804% of the total crude protein in unsprayed and sprayed fuzzy seed, respectively.

Results showed Cry1Ab and PAT analyte proteins were expressed in the kernels, lint coats and, by inference, in the fuzzy seeds derived from the transgenic T304-40 cotton plants. 

CLASSIFICATION:  ACCEPTABLE
                                   47634804
Protein Expression Analysis of Cotton Event T304-40, Expressing Cry1Ab and PAT/bar Proteins, USA, 2007
The purpose of this study was to quantify the expression of Cry1Ab and PAT proteins in transgenic cotton Event T304-40, and in a non-transgenic cotton line Coker 315. Extracts of plant matrices were prepared from roots, stems, leaves, squares, apex, bolls, whole plant, pollen, nectar, flowers and grain or delinted seed. An enzyme-linked immunosorbent assay (ELISA) was used for determining the protein levels.  

On a dry weight basis, the overall averages for the Cry1Ab protein content were 8.6 and 3.8 ug/g for roots at the two growth stages; 6.2 and 0.88 ug/g for stems at the two growth stages; 4.6, 1.4 and 0.45 ug/g for leaves at the three growth stages; 2.9 ug/g for squares; 1.3 ug/g for apex; 0.38 ug/g for bolls; 2.2 ug/g for whole plant; 0.05 ug/g for pollen (fresh weight data only due to insufficient availability of material); 0.002 ug/g for nectar (fresh weight data only due to insufficient availability of material); 10.60 ug/g for flowers and 4.10 ug/g for grain. 

On a dry weight basis, the overall averages for the PAT protein content were reported to be 115 and 95.1 ug/g for roots at two growth stages (ca. 4 weeks and 10 weeks, respectively); 100 and 44.5 (ug/g for stems at two growth stages (ca. 4 weeks and 10 weeks, respectively); 193, 134 and 57.9 ug/g for leaves at 3 growth stages (ca. 4 weeks, 7 weeks, and 10 weeks, respectively); 180 ug/g for squares (ca. 7 weeks); 88.9 ug/g for apex (ca. 10 weeks); 39.3 ug/g for bolls (ca. 8 weeks); 134 ug/g for whole plant (ca. 10 weeks); 0.13 ug/g for pollen fresh weight data only due to insufficient availability of material) (ca. 7 weeks); 0.005 ug/g for nectar fresh weight data only due to insufficient availability of material) (ca. 7 weeks); 187 ug/g for flowers (ca. 10 weeks); and 39.6 ug/g for grain (ca. 22 weeks). 

The recoveries of both proteins from the spiked sample matrices were generally good (means of 59.9-121%) thus validating the use of the ELISA used in this study as a suitable test kit for the detection of both Cry1Ab and PAT proteins as expressed in Event T304-40 cotton.  

CLASSIFICATION:  ACCEPTABLE
                                   47634803
Detailed insert characterization of Gossypium hirsutum transformation Event T304-40
Southern blot analyses were conducted to determine the organization of the cry1Ab and bar gene cassette in transformation Event T304-40 cotton. Genomic DNA extracted from leaves of Event T304-40 cotton were digested with ten different restriction enzymes and the resulting DNA fragments were separated by agarose gel electrophoresis, transferred to a membrane and sequentially hybridized with six different probes (5'e1-Ps7s7 promoter, cry1Ab gene, 3'me1 terminator, P35S3 promoter, bar gene and 3'nos terminator), each representing a fragment of the transforming gene cassette, and with the complete T-DNA probe. 
The Southern blot data for the expected and observed hybridization fragments, as well as the hybridization strategy demonstrated that the inserted transgenic sequence in the cotton Event T304-40 consists of one nearly complete copy of the T-DNA flanked by an inverted incomplete copy of the cry1Ab gene cassette and one additional 3'me1 terminator. 

CLASSIFICATION:  ACCEPTABLE
                                   48480003
                   [This study supersedes MRID No. 47634801]
Confirmation of the absence and presence of
vector backbone sequences in Gossypium hirsutum transformation Event T304-40
The purpose of this study was to determine the presence of any vector backbone sequences in Event T304-40 cotton after insertion of the intended cry1Ab gene cassette via the pTDL008 transformation vector to produce Event T304-40 cotton. Isolated genomic DNA samples from leaves of cotton Event T304-40 and from the non-transgenic negative control were subjected to Southern blot analysis, after digestion with restriction enzymes EcoRV and NotI, using seven overlapping vector backbone probes that encompassed the entire backbone of the pTDL008 transformation vector used to produce this transformation event. The vector backbone probes were then stripped from the membranes so that they could be re-hybridized with the T-DNA probe, as a positive control. Genomic DNA from the Event T304-40 cotton showed no hybridization with any of the vector backbone probes, while the positive control showed hybridization as expected. The results confirmed the absence of vector backbone sequences  after insertion of the intended cry1Ab gene cassette via the pTDL008 transformation vector to produce Event T304-40 cotton.

CLASSIFICATION:  ACCEPTABLE
                                   48480004
                   [This study supersedes MRID No. 47634801]
Characterization of Cry1Ab and PAT/bar proteins produced in Event T304-40 cotton, Gossypium hirsutum
The purpose of this study was to determine the biochemical and functional equivalence of plant-expressed Cry1Ab protein produced in Event T304-40 cotton plants  with bacterial-expressed Cry1Ab protein (Test substances ID No. MIN1443 and AW200308) produced from recombinant Escherichia Coli over-expression system, for its use as a suitable test substance surrogate in toxicological testing. Equivalence studies were also conducted for plant- expressed PAT protein in Event T304-40 plants with bacterial-expressed PAT protein (Test substance ID No. NB010905P44P2) for use as a test substance surrogate. Six analytical tests were used to assess the  equivalence, which included: 1) estimation of molecular weight by observing protein mobility in SDS-PAGE, 2) verification of immunoreactivity using western blot analysis, 3) analysis of biological activity using enzymatic activity for the PAT/bar protein and insecticidal activity for the Cry1Ab protein, 4) analysis of glycosylation using the reaction with glycostain, 5) mass spectroscopic peptide mapping using HPLC/Electrospray Mass Spectrometry (LC/MS) of peptides, and 6) analysis of N-terminal sequencing using Edman degradation. The molecular weights, immunoreactivity and biological activity were shown to be comparable for each respective form of bacterial- and plant-expressed Cry1Ab and PAT proteins. Glycosylation was not detected in either protein from either species. Except for a few amino acids at the N-terminal end, the PAT/bar protein was shown to be identical in both species in the LC/MS analysis. A high degree of similarity was also shown between the Cry1Ab proteins of both species, with sequences between residues 2 and 617 being identical. Slight differences are present in both the N- and C-terminal peptides. Sequencing of the N-terminal end of the Cry1Ab protein using Edman degradation in cotton was unsuccessful, yet consistent with the findings in mass spectroscopic peptide mapping using LC/MS. The comparison between species in the LC/MS assay on the PAT/bar protein showed comparable results and confirmed the biochemical and functional equivalence. Thus, both the bacterial-expressed Cry1Ab (Test substances ID No. MIN1443 and AW200308) and PAT (Test substance ID No. NB010905P44P2)  proteins would be suitable test substance surrogates for use in toxicological testing to  establish the safety of the plant-expressed Cry1Ab and PAT proteins as expressed in Event T304-40 cotton.

CLASSIFICATION:  ACCEPTABLE 
                                   48480005
           [This study supersedes MRIDs No. 47634805 and  47640006]
Protein Detection Methods. Enzyme Linked Immuno Sorbent Assay (ELISA)
Non-quantitative detection of Cry1Ab protein in extracts of plant material from Event T304-40 cotton can accomplished using the EnviroLogix QuantiPlate Kit or the GMO√ Bt Maize Test Kit, each of which uses enzyme-linked immunosorbent assay (ELISA) to detect the presence of Cry1Ab protein.

CLASSIFICATION:  SUPPLEMENTAL pending verification of the specific ELISA test kit method to be used as the proposed residue analytical method for the detection of Cry1Ab protein in Event T304-40 cotton in addition to an independent laboratory validation study to evaluate the designated test kit's performance. 
                                   47634826

   A. III. ACUTE TOXICOLOGY

Cry1Ab Protein

EPA evaluated human health toxicology data for Cry1Ab protein in Event T304-40 as summarized below in Table A.3.

Table A.3. Human Health Data for Event T304-40 cotton expressing Bt Cry1Ab protein
                          (EPA File Symbol 264-RNOU)
                                  Study Title
               A.3. Human health - Summary  -  Bt Cry1Ab protein
                                    MRID No
Cry1Ab protein - Overall amino acid sequence homology search with known toxins and allergens 

The amino acid sequence of the Cry1Ab protein encoded by the plant-optimized Cry1Ab gene from Bt subspecies berliner 1715 was compared with all protein sequences present in the following databases: SwissProt, trEMEL, GeneSeq-Prot, PIR, PDB, DAD, and GenPept. No similarities between the Cry1Ab protein and known allergens or toxins other than Bt toxins were found using the criterion of 35% identity over an 80 amino acid segment.

Classification: ACCEPTABLE
                                   46708802 
                                       
                   [This study supersedes MRID No. 46474402]
Cry1Ab protein -  Epitope homology search.

The amino acid sequence of the Cry1Ab protein was compared to amino acid sequences of proteins in the Allergen database to identify any short sequence similarities with known allergens. No similarities between the Cry1Ab protein and known allergens were identified based on 100% identity over an eight amino acid segment.

Classification: ACCEPTABLE
                                   46708803 
                                       
                   [This study supersedes MRID No. 46474403]
Cry1Ab Protein: Epitope Homology, N-glycosylation and Overall Amino Acid Sequence Homology Search with Known Toxins and Allergens - Supplemental data
Updated studies (conducted in 2008) on amino acid sequence homology of Cry1Ab protein with known toxins and allergens and for N-glyco - sylation sites showed no changes from the previously conducted studies done in 2005 (MRIDs No. 46708802 and 46708803, respectively).




Classification: SUPPLEMENTAL; No additional data needed.
                                   47634809
                                       
     [This study is supplemental data to MRIDs No.  46708802 and 46708803]
Comparative analysis of the Cry1Ab protein amino acid sequence
The amino acid sequence of the Cry1Ab protein encoded by the gene in BCS CropScience's (BCS) Bt cotton plants was compared with the sequence of the wild type Cry1Ab5 protein and the Cry1Ab protein in a previously registered PIP in corn, Bt11. The deduced amino acid sequence of BCS's Cry1Ab protein is identical to the amino acid sequence of the Bt11 Cry1Ab (OPP chemical code 006444) with the exception of one amino acid at the C terminus and two amino acids at the N-terminus. In addition, BCS's Cry1Ab is identical to a portion of the wild type Cry1Ab5 protein produced in Bt subspecies berliner 1715, except for two amino acids at the N-terminus.

Classification: ACCEPTABLE
                                   46708806 
                                       
                   [This study supersedes MRID No. 46474406]
Acute Toxicity by Oral Gavage in Mice- Cry1Ab Protein
The acute oral toxicity of Cry1Ab protein was assessed on a group of five female mice by administering bacterial-produced Cry1Ab protein by oral gavage at a total dose of 2000 mg/kg body weight. Due to the limitation of the solubility of the test material,  the group of 5 female mice were given two doses of 1000 mg/kg body weight in a solution of 50 mM Na2CO3 (pH 9.6) within  a 4 hour time period on the day of treatment. Animals were observed for 15 days. At test end, there were no mortalities, and no treatment-related adverse effects observed. Therefore, the acute oral LD50 of the Cry1Ab protein is greater than 2000 mg/kg body weight, which places Cry1Ab protein in EPA TOXICITY CATEGORY III for acute oral exposure.

CLASSIFICATION:  ACCEPTABLE
                                   47651103
Cry1Ab Protein: Heat stability study
An in vitro heat stability study was conducted on bacterial-expressed Cry1Ab protein at temperatures of 60, 75 or 90°C for incubation periods of 10, 30 or 60 minutes and examined by Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE) and by Western blot analysis. Results of the SDS-PAGE and Western blot analyses showed the Cry1Ab protein had no significant changes in protein stability after heat treatment at 60°C for 10, 30 and 60 minutes and at 75°C for 10 and 30 minutes. After incubation for 60 minutes at 75°C, the Cry1Ab protein was slightly decreased. For the 90°C treatment, denaturation was observed after 10 and 30 minutes, and finally at 60 minutes, a marked but not complete degradation of the Cry1Ab protein was observed after incubation at 90°C. Additional protein bands of lower molecular weight (mostly likely representing degradation products) also appeared as the ~69 kDa Cry1Ab protein band disappeared. Results showed consistent patterns of decreasing structural stability of the Cry1Ab protein in between both analyses tested and was concluded as partially heat-stable at 90°C for 60 minutes.

CLASSIFICATION:  ACCEPTABLE
                                   47651104
Cry1Ab Protein: In vitro digestibility study in human simulated gastric fluid
An in vitro digestibility study was conducted on bacterial-expressed Cry1Ab protein in simu - lated human gastric fluid containing pepsin (pH 1.2) to assess the susceptibility of Cry1Ab protein to proteolytic degradation. SDS-PAGE analysis of the Cry1Ab protein solution showed the expected molecular weight of ~69 kDa at the initial time point. After incubation in SGF for 2 minutes, the band for the Cry1Ab protein was no longer visible on the Coomassie blue stained gel. This study indicates that the Cry2Ae protein is rapidly digested in SGF containing pepsin at pH 1.2.

CLASSIFICATION:  ACCEPTABLE
                                   47651102
Cry1Ab Protein: In vitro digestibility study in simulated intestinal fluid
An in vitro digestibility study was conducted on bacterial-expressed Cry1Ab protein in simu - lated human intestinal fluid (SIF) containing pancreatin (pH 7.5) to assess the susceptibility of Cry1Ab protein to proteolytic degradation. The Cry1Ab protein was incompletely degraded into several stable fragments in simulated intestinal fluid containing pancreatin  (ph 7.5).

CLASSIFICATION:  ACCEPTABLE
                                   47651101
ORF Analysis -Toxicology (Human Health Assessment)
The purpose of this study was to evaluate newly created ORFs that span the 5' or 3' junction of the transgenic insert (T-DNA) in cotton Event T304-40 or the two internal junctions that were created within the T-DNA insert during the transformation process. ORFs were identified using the GetORF search program from the EMBOSS (European Molecular Biology Open Software Suite) tools. Two ORFs of interest were found. No significant similarities were identified between the ORF-1 putative amino acid sequence and sequences from the databases. The ORF-2 putative amino acid sequence was identical to a fragment of the polymerase from the rice ragged stunt virus (RRSV), with a 100% match over the 28 putative amino acids generated by this ORF. Since the putative ORF-2 sequence is only 28 amino acids long, while the RRSV polymerase has 1357 amino acids, it is unlikely that the ORF-2 polypeptide has any polymerase activity even if it was expressed. In addition, no records were on potential allergenicity or toxicity in animals associated with this virus polymerase protein or with the RRSV organism. Overall, neither of the two putative ORF amino acid sequences of the cotton transformation Event T304-40 generated any significant amino acid sequence similarities with known toxins and known allergens.

CLASSIFICATION:  ACCEPTABLE
                                   47634806




REFERENCES  -  Cry1Ab Protein -  Product Characterization and Human Health 
Höfte H., de Greve H., Seurinck J., Jansens S., Mahillon J., Ampe C., Vandekerckhove J., 
      Vanderbruggen H., van Montagu M., Zabeau M.,Vaeck M. (1986) Structural and functional analysis of a cloned delta endotoxin of Bacillus thuringiensis berliner 1715. Eur. J. Biochem. 161:273-280
Thompson C., Van Montagu M. and Leemans J. (1987) Engineering herbicide resistance in plants by expression of a detoxifying enzyme. EMBO Journal. 6(9): 2513-2518
U.S. EPA. 2006.The product characterization and human health data submitted in support of the EUP. Memorandum from R. Edelstein, through J. Kough, to S. Matten dated January 27, 2006. US Environmental Protection Agency. Washington DC
U.S. EPA. 2011. Review of Product Characterization Data for the Registration of Plant-Incorporated Protectant (PIP) cotton Events: GHB119 cotton expressing Bacillus thuringiensis Cry2Ae protein (EPA File Symbol 264-RNOU), T304-40 cotton expressing Bt CrylAb (EPA File Symbol 264-RNOL); and TwinLink(TM) cotton (EPA File Symbol 264-RNOA), its associated combination PIP product developed by conventional breeding of cotton Events GHB119 and T304-40.  Memorandum from A. Waggoner, through J. Kough, to S. Bacchus and D. Greenway dated February 4, 2011. US EPA. Washington DC
U.S. FDA. 1992. Statement of Policy: Foods Derived From New Plant Varieties. Federal Register: Vol. 57 No. 104 p. 22984. Dated: of May 29, 1992. Department Of Health And Human Services- Food and Drug Administration (Docket No. 92N-0139) http://www.cfsan.fda.gov/~lrd/biotechm.html#reg
Van Rie J. 2004. Comparative analysis of the Cry1Ab protein amino acid sequence. Project Number: None; Unpublished study prepared by Bayer CropScience LP. MRID No. 46474406
APPENDIX B:  Product characterization and human health effects - Bacillus thuringiensis Cry 2Ae (Event GHB 119) and TwinLink(TM) in cotton plant - incorporated protectant.

--------------------------------------------------------------------------------
Event GHB119 Cotton (OECD Unique Identifier BCS-GH5-8) Expressing Cry2Ae protein

Appendix B contains summaries of EPA's reviews of the product characterization and mammalian toxicology of submitted studies for Bacillus thuringiensis Cry 2Ae and TwinLink(TM) (combined Cry1Ab x Cry2Ae proteins).

BCS has developed  Event GHB119 cotton line expressing Cry2Ae insecticidal protein derived from Bacillus thuringiensis (Bt), which is intended to control certain lepidopteran pest species such as bollworm (CBW, Helicoverpa zea), pink bollworm (PBW, Pectinophora gossypiella), tobacco budworm (TBW, Heliothis virescens) larvae, and fall armyworm (FAW, Spodoptera frugiperda). In 2004, BCS submitted product characterization and human health data for the purposes of conducting time-limited experimental field trials with GHB119 cotton as a new PIP and a petition for a temporary exemption from the requirement of a tolerance of Cry2Ae protein residues on food/feed commodities in cotton. The EUP also included testing TwinLink(TM) cotton (EPA File Symbol 264-RNOA), a new combination PIP product, developed by conventional breeding of the parental Events T304-40 x GHB119. 

The product characterization and human health data submitted for Event GHB119 and TwinLink(TM) cotton were reviewed and found acceptable for the EUP application (US EPA, 2008). As part of that regulatory action, the Agency also conducted a preliminary food safety assessment for the Cry2Ae protein expressed in Event GHB119 cotton and a temporary tolerance exemption was granted for Cry2Ae protein residues produced in this PIP cotton line (40 CFR § 174.530). Event GHB119 cotton also expresses phosphinothricin acetyltransferase (PAT) enzyme, which is exempt from the requirement of a tolerance in all food commodities when used as a PIP inert ingredient (40 CFR § 174.522).  

In May 2009, BCS submitted applications for FIFRA section 3 registrations for PIP Events GHB119, T304-40 and TwinLink(TM) cotton, their associated combination PIP product developed by traditional breeding methods. Concurrently, BCS filed a petition to establish a permanent tolerance exemption for Cry2Ae protein in all food commodities on the basis of the submitted data (see REGULATORY BACKGROUND, Chapter II.B.2.b). These data were reviewed by the Agency and were found supplemental. Several product characterization studies and additional data were required by the Agency to address the data deficiencies identified in those studies (US EPA, 2011a).  

In May 2011, BCS resubmitted studies to address the data deficiencies. EPA reviewed the additional data submitted for Event GHB119 (Cry2Ae) cotton and determined they are acceptable for registration of the PIP and for establishing the permanent tolerance exemption for Cry2Ae expressed in Event GHB119 (US EPA 2011c). Summaries of EPA's evaluation of product characterization and the remaining toxicological data for Cry2Ae PIP are available in this Appendix.

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
FFDCA FOOD SAFETY ASSESSMENT
--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
Event GHB119 Cotton (OECD Unique Identifier BCS-GH5-8) Expressing Cry2Ae protein

   B. I. PRODUCT CHARACTERIZATION

BCS developed Event GHB119 cotton line expressing Cry2Ae insecticidal protein derived from Bacillus thuringiensis (Bt), which is intended to control certain lepidopteran pest species such as bollworm (CBW, Helicoverpa zea), pink bollworm (PBW, Pectinophora gossypiella), tobacco budworm (TBW, Heliothis virescens) larvae, and fall armyworm (FAW, Spodoptera frugiperda).

a. Transformation System

Event GHB119 cotton was produced by Agrobacterium tumefaciens-mediated transformation of Gossypium hirsutum L. cultivar Coker 312  using transformation vector pTEM12 carrying T-DNA containing the cry2Ae gene, which encodes for Cry2Ae protein and confers insect resistance to certain lepidopteran pest species. The cry2Ae gene was isolated from Bacillus thuringiensis subspecies dakota and its sequence modified for optimal expression in plants. A sequence encoding a 58 amino acid transit peptide from the small subunit of the rubisco gene from Arabidopsis thaliana (TpssuAt) was added at the 5' end of the coding region of the cry2Ae gene. BCS states that the presence of this transit peptide targets the Cry2Ae protein to the chloroplasts in the transformed cells and that the transit peptide is cleaved off during the transfer of the protein through the choloroplast membrane. Plasmid pTEM12 also contains the bar gene as a selectable marker, which is derived from the soil microorganism, Streptomyces hygroscopicus (described by Thomson et al. 1987). The bar gene encodes the enzyme phosphinothricin acetyltransferase (PAT), which is co-expressed with Cry2Ae protein in Event GHB119 cotton as a PIP inert ingredient. The PAT enzyme provides the plant with tolerance to herbicide applications containing glufosinate-ammonium by acetylating the herbicide glufosinate-ammonium and making it unable to bind glutamine synthase. The PAT enzyme encoded by the bar gene is a homodimer of 183 amino acids and has an expected molecular weight of approximately 22 kDa. 

b. Characterization of the DNA Inserted in the Plant and Inheritance and Stability

Characterization of the DNA isolated from Event GHB119 cotton using restriction enzyme digests and Southern blot analyses as well as DNA sequencing indicates that the DNA was inserted in the cotton genome at a single locus, and the insert contains one copy of the cry2Ae gene. No vector backbone sequences from the transformation plasmid pTEM12 were found in GHB119. Inheritance and stability studies of the cry2Ae gene in GHB119 verified that it is stably integrated into the cotton genome, segregating in an expected Mendelian fashion.

c. Protein Characterization 

Event GHB119 cotton expresses Cry2Ae protein produced by Bacillus thuringiensis subsp. dakota 1715. The Cry2Ae protein after cleavage of the transit peptide has 631 amino acids. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analyses of Cry2Ae protein as expressed in Event GHB119 showed a molecular weight at approximately 71 kDa. Western blot analyses showed a single immunoreactive band at ~71 kD. Cry2Ae protein was also found to be non-glycosylated. N-terminal sequencing of the plant expressed form also confirmed that the transit peptide of the plant form was cleaved at the expected residue. The biological activity of Cry2Ae protein was assessed by an insect feeding bioassay using Heliothis virescens first instar larvae and the LC50 for the plant- produced Cry2Ae protein was 296 ng/cm².   

d. Protein Expression

Expression level data were provided for Cry2Ae in different plant tissues and at different growth stages in Event GHB119 cotton. A summary of the results obtained using an ELISA method is provided in Table B.1 below. 
                                       
    Table B.1. Mean Cry2Ae Protein Expression levels in Event GHB119 Cotton
                                       
                                  Tissue Type
                                Cry2Ae Protein 
                           (ug/g dry weight +- SD)*
Roots
                         11.2 +- 2.4  -   16.9 +- 2.8
Stems
                        5.26 +- 0.87  -   5.58 +- 1.72
Leaves
                        17.8 +- 7.1   -   37.5 +- 13.6
Squares
                                 8.69 +- 3.72
Apex
                                 8.38 +- 2.29
Bolls
                                 4.13 +- 0.80
Whole plants
                                 30.4 +- 13.2
Pollen 
                                    ND [a]
Nectar 
                                    NA [b]
Flowers
                                  5.59+- 0.75
Cottonseed
                                 0.99 +- 0.61
		*Range reflects means at different growth stages 
                     a Not detected.
					[b] Not applicable due to the insufficient availability of material.

e. Supporting Data

The product characterization studies to support the applicant's registration of PIP cotton Event GHB119 are summarized in Table 2. A number of studies were reviewed under a work-share program according to the EPA-BPPD and USDA-BRS memorandum of understanding (09-2000-0052-MU). For detailed information of each study and secondary reviewer information, please refer to the individually reviewed studies in the Data Evaluation Reports (DERs) according to its respective assigned MRID number).  
Table B.2. Product Characterization Data for Event GHB119 cotton expressing Bt Cry2Ae protein
                          (EPA File Symbol 264-RNOL)

                                  Study Title
              B.2. Summary- Cry2Ae PIP - Product Characterization
                                    MRID No
Characteristics of Cry2Ae Cotton Plants Derived from Transformation Events GHB119 and GHB714: Preliminary Report and Supplemental Information
This submission includes descriptions of the genetic construct used in Events GHB119 and GHB714, the sources of the introduced genes, the modes of action of the expressed proteins, and preliminary molecular analysis. Restriction enzyme digestion of genomic DNA from leaf material from Events GHB119 and GHB714 cotton and hybridization with the complete T-DNA probe gave the expected results, indicating that the Events contain a single, intact copy of the expected T-DNA. 

CLASSIFICATION:  ACCEPTABLE
                                   47125101
PAT and Cry2Ae Protein in Cotton Tissues of Events GHB119 and GHB714. Preliminary Report[1]
BCS conducted preliminary protein expression analysis for Cry2Ae cotton. Average Cry2Ae protein concentrations, as determined by ELISA in ground cotton seed from Events GHB119, GHB714, and GHB866 were 3.5 +- 0.9 g/g, 2.9 +- 0.8 g/g, and 1.9 +- 0.3 g/g fresh weight, respectively. Average PAT concentrations, as determined by ELISA in ground cotton seed from Events GHB119, GHB714, and GHB866 were 114 +- 25 g/g, 131 +- 23 g/g, and 100 +- 20 g/g, fresh weight, respectively. The percent dry weight equivalents in seeds are: Cry2Ae: 0.0002-0.00037% and PAT: 0.01- 0.0 14%. Leaf, boll, square, and flower samples were also taken at various growth stages and analyzed for Cry2Ae from plants grown both in the field and in a greenhouse. The combined results from all three Events gave the following ranges on a percent total soluble protein basis:  leaf: 0.026 to 0.068%; boll: 0.007 to 0.036%; square: 0.008 to 0.02 1%; and flower: 0.018 to 0.02 1%. The PAT content of cotton tissues (other than seed) was not presented in the study report. 

Classification: SUPPLEMENTAL; Sufficient to support EUP; Data gaps were addressed for Sec. 3 registration in MRID No. 47641903 

                                   46708908
Product Characterization of Cry2Ae Cotton Event GHB119
The purpose of this study was to determine if the transfer DNA (T-DNA) containing the intended cry2Ae gene from plasmid pTEM12 was inserted into the genomic DNA of cotton Event GHB119 and the backbone DNA from pTEM12 was not inserted into cotton Event GHB119. Southern blot analyses demonstrated a single copy of the intended cry2Ae gene from plasmid pTEM12 was inserted into the genomic DNA of cotton Event GHB119 and the configuration of the inserted DNA corresponds to that of the original transformation vector. Results also demonstrated the absence of most of the vector backbone DNA from pTEM12 after transformation Event GHB119

However, there were several data missing from this report and one region of the vector backbone sequence was not confirmed as absent after the transformation process to create GHB119 cotton. Additional analyses such as demonstrating the cry2Ae gene was stably integrated and segregated according to Mendelian laws of inheritance over several successive generations of Event GHB199 cotton were not addressed in this study.

Additional data are needed to complete the molecular characterization of Event GHB119. These include:  a plasmid map with all the enzymatic cleavage sites; a comparison of the expected DNA base pairs of each restriction enzyme digest to the actual fragment sizes observed, confirmation of the absence of all regions in the vector backbone sequence after transformation; and demonstration of the integration and stability of the cry2Ae gene over several successive generations of Event GHB1199 cotton.

CLASSIFICATION:  SUPPLEMENTAL; Data gaps were addressed for Sec. 3 registration in MRID No. 48480002 and 48480003
                                   47641901
Structural and Functional Equivalence of Cry2Ae and PAT/bar  Proteins Produced in Bacillus thuringiensis (Bt) and Escherichia coli to Cry2Ae and PAT/bar Proteins in GHB119 Cotton, Gossypium hirsutum[2]
The purpose of this study was to demonstrate the biochemical and functional equivalence of the Cry2Ae and PAT/bar proteins expressed in cotton Event GHB119 with the Cry2Ae protein expressed in Bacillus thuringiensis (Bt) and the PAT/bar protein expressed in Escherichia coli (E. coli), and the Cry2Ae and PAT/bar proteins expressed in TwinLink(TM) cotton. The results from SDS-PAGE analyses showed that the Cry2Ae protein isolated from Event GHB119 cotton leaves, the Cry2Ae E. coli protein standard, and TwinLink(TM) cotton had identical electrophoretic mobilities and similar molecular weights (~66 kDa). SDS-PAGE analyses also showed identical electrophoretic mobilities and similar molecular weights (~20 kDa) for the PAT/bar protein isolated from Event GHB119 cotton leaves, from the PAT/bar E. coli standard and the TwinLink(TM) cotton. However, it is unclear in the study report if the biochemical data presented in the study were derived from the same source of the bacterial-expressed Cry2Ae protein test substance. The study report lacked sufficient details on the materials/ methods used, the experimental design, and reported results to properly document the bioequivalence of the bacterial-derived Cry2Ae protein test substance to the T304-40 cotton plant-expressed Cry2Ae protein. Several single-dose sensitive insect bioassays were also conducted to confirm bioactivity, but these data did not provide the quantitative results to allow an adequate comparison of the insecticidal activity for each source of protein. Therefore, the biochemical and functional equivalence of the plant- and bacterial- expressed protein sources cannot be confirmed from these data.    

An additional sensitive insect bioassay study is required to characterize the insecticidal activity and establish the dose-response relationship of Cry2Ae protein against the target pest. Additional information is also needed to verify the validity of the results that were reported for Cry2Ae bacterial-derived protein substances to confirm that samples were tested from the same original source of preparation.   

CLASSIFICATION:  SUPPLEMENTAL; Data gaps addressed in MRID No. 48480003 
                                  47641905   
Analyses of Raw Agricultural Commodity (Fuzzy Seed) of Cry2Ae Cotton Event GHB119 for PAT/bar and Cry2Ae and its Non-transgenic Counterpart for PAT/bar and Cry2Ae Proteins[2]
The purpose of this study was to quantify the expression of Cry2Ae and PAT proteins in the fuzzy seeds of transgenic cotton Event GHB119 and in a non-transgenic cotton line, Coker 312. Fuzzy seed was ground into fractions of lint coats and kernels and analyzed separately for Cry2Ae and PAT proteins and total extractable protein. To quantitate the amount of the protein analytes in the fuzzy seeds, ELISA results of separate assays were combined for the kernel and lint coat fractions. 
The average Cry2Ae protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, on a dry weight basis was 1.55 +- 0.19 ug/g and 1.47 +- 0.07 ug/g, respectively. As a percent of total crude protein, Cry2Ae protein comprised an average of 0.00084 +- 0.00010% and 0.00082 +- 0.00007% of the total crude protein in unsprayed and sprayed fuzzy seed, respectively. The average PAT protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, on a dry weight basis, was 50.7 +- 4.1 ug/g and 49.9 +- 3.5 ug/g, respectively. As a percent of total crude protein, PAT protein comprised an average of 0.027 +- 0.002% and 0.028 +- 0.003 % of the total crude protein in unsprayed and sprayed fuzzy seed, respectively.

Overall results showed Cry2Ae and PAT analyte proteins were expressed in the kernels, lint coats and, by inference, in the fuzzy seeds derived from the transgenic GHB119 cotton plants.  

CLASSIFICATION:  ACCEPTABLE
                                   47641904
Protein Expression Analysis of Cotton Event GHB119, Expressing Cry2Ae and PAT/bar Proteins, USA, 2007
The purpose of this study was to quantify the expression of Cry2Ae and PAT proteins in transgenic cotton Event GHB119 and in a non-transgenic cotton line Coker 312. Extracts of plant matrices were prepared from the roots, stems, leaves, squares, apex, bolls, whole plant, pollen, nectar, flowers and cottonseed. An enzyme-linked immunosorbent assay (ELISA) was used for determining the protein levels. There was a range of extraction efficiencies for each plant matrix and for each protein (means of 64.2-142%). On a dry weight basis, the overall averages for the Cry2Ae protein content were 16.9 and 11.2 ug/g for roots at two growth stages; 5.58 and 5.26 ug/g for stems at two growth stages; 37.5, 33.3, and 17.8 ug/g for leaves at three growth stages; 8.69 ug/g for squares; 8.38 ug/g for apex; 4.13 ug/g for bolls; 30.4 ug/g for the whole plant; not analyzed for pollen due to the insufficient availability of material, non-detectable for nectar, 5.59 ug/g for flowers, and 0.99 ug/g for cottonseed. Generally, the average PAT protein concentrations reported in the plant matrices from GHB119 were ca. an order of magnitude greater than the Cry2Ae protein concentrations. This study showed that quantifiable levels of Cry2Ae and PAT proteins were found in all plant matrices taken from the transgenic cotton Event GHB119 (except nectar, where only PAT protein was detected), while these proteins were absent from all of the cotton matrices derived from the non-transgenic cotton Coker 312. Pollen and nectar contained the least amount of both Cry2Ae and PAT proteins.
   
CLASSIFICATION:  ACCEPTABLE
                                   47641903
Protein Detection Methods. Enzyme Linked Immuno Sorbent Assay (ELISA)
Semi- and full quantitative detection of Cry2Ae protein in extracts of plant material from Event GHB119 cotton can be accomplished using the EnviroLogix QuantiPlate Kit for Cry2A, which uses enzyme-linked immunosorbent assay (ELISA) analyses to detect the presence of Cry2A protein residues.

CLASSIFICATION:  SUPPLEMENTAL- An independent laboratory validation study to evaluate the ELISA test kit's performance as the designated analytical method for the detection of Cry2Ae protein residues is needed. 
                                   47641926
Product Characterization  - 
Nutritional Characterization of Cry2Ae cotton Event GHB119
The purpose of this study was to perform a compositional analysis of nutritional factors, including protein, fat, carbohydrate, minerals, vitamins, amino acids and fatty acids (cottonseed oil), and anti-nutritional factors, including gossypol, phytic acid and cyclic fatty acids present in transgenic cotton Event GHB119 (without and with herbicide treatment), and to compare these nutritional values to those found in a non-transgenic cotton line (Coker 315) and to nutritional values generally found in cottonseed reported from other sources.


CLASSIFICATION:  SUPPLEMENTAL; Data not required for EPA Sec.3 Registration of PIPs. The nutritional and compositional analysis of Event GHB119 cotton falls under the regulatory purview of FDA.

                                   47641902
Detailed insert characterization of Gossypium hirsutum transformation Event GHB119 by Southern blot analysis
Southern blot analyses were conducted to determine the organization of the cry2Ae and bar gene cassette in transformation Event GHB119 cotton. Genomic DNA extracted from leaves of Event GHB119 cotton was digested with one of the following restriction enzymes or enzyme complexes: DraI, EcoRV, HindIII, NcoI, BamHI, EcoRI, PstI, AvaI, NdeI, XhoI, EcoRI/PstI and EcoRI/NdeI). The resulting DNA fragments were separated by agarose gel electrophoresis, transferred to a membrane and sequentially hybridized with six different probes cassette (3'nos, bar, Pcsvmv, P35S2-5'cab22L, TPssuAT-cry2Ae and 3'35S-RB), each representing a fragment of the transforming gene cassette, and with the complete T-DNA probe. The Southern blot data for the expected and observed hybridization fragments, as well as the hybridization strategy demonstrated that the inserted transgenic sequence in the cotton Event GHB119  is integrated as a single copy in Event GHB119 cotton. Based on plant breeding activities, analyses from breeding and backcrossing activities with GHB119 generations, the Chi-square analyses of the ratio of expected vs. observed F1 and F2 progeny demonstrated that the cry2Ae gene insertion was stably integrated and segregated according to Mendelian laws of inheritance.
 
CLASSIFICATION:  ACCEPTABLE
                                   48480001
                                       
                   [This study supersedes MRID No. 47641901]
Confirmation of the absence of vector backbone sequences in Gossypium hirsutum transformation Event GHB119
The purpose of this study was to determine the presence of any vector backbone sequences in Event GHB119  cotton after insertion of the intended cry2Ae gene cassette via the pTEM12 transformation vector. Isolated genomic DNA samples from leaves of cotton Event GHB119  and from the non-transgenic negative control were subjected to Southern blot analysis, after digestion with restriction enzymes EcoRV and NotI, using seven overlapping vector backbone probes that encompassed the entire backbone of the pTEM12 transformation vector used to produce this transformation event. The vector backbone probes were then stripped from the membranes so that they could be re-hybridized with the T-DNA probe, as a positive control. Genomic DNA from the Event GHB119 cotton showed no hybridization with any of the vector backbone probes, while the positive control showed hybridization as expected. The results confirmed the absence of vector backbone sequences after insertion of the intended cry2Ae gene cassette via the pTEM12 transformation vector to produce Event GHB119 cotton.

CLASSIFICATION:  ACCEPTABLE
                                   48480002
                                       
                   [This study supersedes MRID No. 47641901]
Characterization of Cry2Ae and PAT/bar proteins expressed in Gossypium hirsutum transformation Event GHB119
The purpose of this study was to determine the biochemical and functional equivalence of plant-expressed Cry2Ae protein produced in Event GHB119cotton plants with bacterial-expressed Cry2Ae proteins (Batch Nos. VMLV1041-1 and NB210806P25 produced from recombinant Bacillus thuringiensis  and  Escherichia coli over-expression system, respectively) for their use as suitable test substance surrogates in toxicological testing. Equivalence studies were also conducted for plant- expressed PAT protein in Event GHB119 plants with bacterial-expressed PAT protein (Batch Nos. NB010905P44P2) for use as a test substance surrogate. Six analytical tests were used to assess the biochemical and functional  equivalence, which included: 1) estimation of molecular weight by observing protein mobility in SDS-PAGE, 2) verification of immunoreactivity using Western blot analysis,  3) analysis of glycosylation using the reaction with glycostain, 4) mass spectroscopic peptide mapping using HPLC/Electrospray Mass Spectrometry (LC/MS) of peptides, 5) analysis of N-terminal sequencing using Edman degradation, and , 3) analysis of biological activity using a sensitive insect bioactivity feeding assay to determine the  insecticidal activity for the Cry2Ae proteins and  enzymatic activity for the PAT/bar protein,

The Cry2Ae protein purified from GHB119 cotton plants and the bacterial-produced batches were found to have a similar molecular weight after electrophoresis, similar immunoreactive bands based on Western Blot analyses and non-glycosylated. Identity of the plant-produced Cry2Ae protein and the bacterial-produced protein batches was confirmed by peptide mapping using electrospray mass spectrometry. After CNBr treatment and tryptic digest, 66% of the plant Cry2Ae protein, 73% of bacterial-produced batch VMLV1041-1 and 75% of bacterial-produced batch NB210806P25 were identified. The N-terminal sequence of the Cry2Ae protein batches was confirmed by Edman degradation. N-terminal sequencing also confirmed that the transit peptide of the plant form was cleaved at the expected residue. Biological activity was assessed by an insect bioassay using Heliothis virescens first instar larvae and the obtained LC50 for the plant-produced Cry2Ae protein was 296 ng/cm[2], while the bacterial-produced Cry2Ae proteins from Batch Nos. VMLV1041-1 and NB210806P25 were 311 ng/cm[2] and 346 ng/cm[2], respectively, demonstrating that the plant- and bacterial-produced proteins have comparable biological activity. 

Thus, both the bacterial-expressed Cry2Ae (Batch No. VMLV1041-1 and NB210806P25) and PAT (Batch No. NB010905P44P2) proteins are suitable test substance surrogates for use in toxicological testing for establishing the safety of the plant-produced Cry2Ae and PAT proteins as expressed in Event GHB119 cotton.

CLASSIFICATION:  ACCEPTABLE 
                                   48480006
                                       
            [This study supersedes MRID No. 47641905 and 47641906]

B. II.  	HUMAN HEALTH ASSESSMENT  -   Cry2Ae PIP

Section 408(c)(2)(A)(i) of the Federal Food, Drug, and Cosmetic Act (FFDCA) allows EPA to establish an exemption from the requirement for a tolerance (the legal limit for a pesticide chemical residue in or on a food) only if EPA determines that the exemption is "safe." Section 408(c)(2)(A)(ii) of the FFDCA defines "safe" to mean that "there is a reasonable certainty that no harm will result from aggregate exposure to the pesticide chemical residue, including all anticipated dietary exposures and all other exposures for which there is reliable information." This includes exposure through drinking water and in residential settings, but does not include occupational exposure. Pursuant to section 408(c)(2)(B), in establishing or maintaining in effect an exemption from the requirement of a tolerance, EPA must take into account the factors set forth in section 408(b)(2)(C), which require EPA to give special consideration to exposure of infants and children to the pesticide chemical residue in establishing a tolerance and to "ensure that there is a reasonable certainty that no harm will result to infants and children from aggregate exposure to the pesticide chemical residue... ." 

Additionally, section 408(b)(2)(D) of the FFDCA requires that the Agency consider "available information concerning the cumulative effects of a particular pesticide's residues" and "other substances that have a common mechanism of toxicity." EPA performs a number of analyses to determine the risks from aggregate exposure to pesticide residues. First, EPA determines the toxicity of pesticides. Second, EPA examines exposure to the pesticide through food, drinking water, and through other exposures that occur as a result of pesticide use in residential settings. 

a.  Toxicological Profile

BCS submitted a pesticide tolerance petition (PP 9F7514) under the FFDCA, requesting 40 CFR Part 174 be amended  by establishing an exemption from the requirement of a tolerance for residues of Bacillus thuringiensis (Bt) Cry2Ae insect control protein and the genetic material necessary for its production in or on all food commodities. The toxicological profile of the protein was previously described in the Federal Register of September 10, 2008 (73FR 52591: FRL-8380-1) establishing a temporary tolerance exemption for Cry2Ae protein in or on cotton food/feed commodities (40 CFR §174.530). The data reviewed to support the petition for establishing a permanent tolerance exemption are based upon the biochemical and functional characteristics of Bt Cry2Ae protein and its expression analyses in cotton. No data were submitted for use of this protein as a PIP in other crops nor do any currently approved PIP Events express Cry2Ae protein. Since the submitted data are specific to Cry2Ae protein expressed in cotton, the exemption from the requirement of a tolerance of Cry2Ae protein residues may only be supported in or on cotton food/feed commodities.

Consistent with section 408(b)(2)(D) of the FFDCA, EPA has reviewed the available scientific data and other relevant information in support of this action and considered its validity, completeness and reliability, and the relationship of this information to human risk. EPA has also considered available information concerning the variability of the sensitivities of major identifiable subgroups of consumers, including infants and children.

1) Mammalian Toxicity and Allergenicity Assessment  -  Cry2Ae PIP

BCS submitted acute oral toxicity data to demonstrate the lack of mammalian toxicity at high levels of exposure to the pure Cry2Ae protein (MRID No. 47076902). The acute oral toxicity of Cry2Ae protein was assessed by orally dosing (via gavage) to five female mice with 2,000 milligrams/kilograms bodyweight of purified bacterial-produced Cry2Ae protein test substance (Batch No. NB210806P25,  >93% purity). All treated animals gained weight and had no test material-related clinical signs and no test material-related findings at necropsy. The acute oral LD50 of the Cry2Ae protein is greater than 2000 mg/kg body weight. The Cry2Ae protein does not appear to cause any significant adverse effects at an exposure level of up to 2000 mg/kg bodyweight and supports the finding that the Cry2Ae protein would be non-toxic to mammals.

These data demonstrate the safety of the product at a level of 2000 mg/kg b.w. that is well above maximum possible ppm exposure levels that are reasonably anticipated in the crop (see Table B.1. above). Basing this conclusion on acute oral toxicity data without requiring further toxicity testing and residue data is similar to the Agency position regarding toxicity testing and the requirement of residue data for the microbial Bacillus thuringiensis products from which this plant-incorporated protectant was derived (see 40 CFR §§ 158.2130(d)(1)(i) and 158.2140(d)(7)). For microbial products, further toxicity testing and residue data are triggered by significant adverse acute effects in studies such as the acute oral toxicity study, to verify the observed adverse effects and clarify the source of these effects (Tiers II & III). Since there were no adverse effects in the Tier I acute oral study, further Tiers II and III toxicity testing and residue data were not required.

2) Allergenicity Assessment

When proteins are toxic, they are known to act via acute mechanisms and at very low dose levels (Sjoblad et al. 1992). Since no acute effects were shown to be caused by Cry2Ae protein, even at relatively high dose levels, the Cry2Ae protein is not considered toxic. Further, amino acid sequence comparisons showed no similarities that would raise a safety concern between the Cry2Ae protein and known toxins in protein databases.

Since Cry2Ae is a protein, allergenic sensitivities were considered. Currently, no definitive tests exist for determining the allergenic potential of novel proteins. Therefore, EPA uses a weight-of- evidence approach where the following factors are considered: source of the trait; amino acid sequence similarity with known allergens; prevalence in food; and biochemical properties of the protein, including in vitro digestibility in simulated gastric fluid (SGF) and glycosylation (as recommended by Codex Alimentarius Commission, 2003). Current scientific knowledge suggests that common food allergens tend to be resistant to degradation by acid and proteases; may be glycosylated; and present at high concentrations in the food. 

Summaries of the allergenicity assessment for Cry2Ae protein include: 

      1. Source of the trait. Bacillus thuringiensis, the microorganism from which Cry2Ae protein is derived, is not considered to be a source of allergenic proteins (MRID No. 47125101). 
2. Amino acid sequence. A comparison of the amino acid sequence of Cry2Ae with known allergens showed no overall sequence similarity using the criterion of 35% identity over an 80 amino acid segment or identity at the level of eight contiguous amino acid residues (MRIDs No. 47641908 and 47641909).
      3. Prevalence in food. Protein expression level analyses of Cry2Ae protein confirm its presence at relatively low levels. Dietary exposure is extremely limited and expected to be correspondingly low. Expression in Event GHB119 ranged from 0.99 to 1.55 ug/g or ppm for cottonseed. Thus, the expression has been shown to be in the parts per million range (MRID No. 47641903).
      4. Digestibility. The Cry2Ae protein was rapidly digested within 30 seconds in simulated mammalian gastric fluid containing pepsin at a pH of 1.2 (MRID No. 47125102).  
5. Glycosylation. The Cry2Ae protein expressed in cotton was shown not to be glycosylated. (MRIDs No. 48471901 and 48480006).
      
Considering all of the available information, EPA concluded that the potential for Cry2Ae protein to be a food allergen is minimal. 
      
b. Aggregate Exposures 

Pursuant to FFDCA section 408(b)(2)(D)(vi), EPA considers available information concerning aggregate exposures from the pesticide residue in food and all other non-occupational exposures, including drinking water from ground water or surface water and exposure through pesticide use in gardens, lawns, or buildings (residential and other indoor uses). 

The Agency considered available information on the aggregate exposure levels of consumers (and major identifiable subgroups of consumers) to the pesticide chemical residue and to other related substances. These considerations include dietary exposure under the tolerance exemption and all other tolerances or exemptions in effect for the plant-incorporated protectants chemical residue, and exposure from non-occupational sources. Exposure via the skin or inhalation is not likely since the plant- incorporated protectant is contained within plant cells, which essentially eliminates these exposure routes or reduces these exposure routes to negligible. The amino acid homology assessment included similarity to known aeroallergens. It has been demonstrated that there is no evidence of occupationally related respiratory symptoms, based on a health survey on migrant workers after exposure to Bt pesticides (Bernstein et al. 1999). Exposure via residential or lawn use to infants and children is also not expected because the use sites for the Cry2Ae protein are all agricultural for control of specific insect pest species. Although cotton is not a directly consumed food commodity, humans may be potentially exposed to extremely low levels in the diet, from ingestion of processed cotton products (e.g. cottonseed flour and oil) and, potentially, drinking water. 

The oral toxicity study conducted at a dose of 2,000 mg/kg b.w. showed no adverse effects. Cottonseed may be processed to edible food commodities, cottonseed flour and cottonseed oil. In cottonseed from the parental strain and TwinLink(TM), Cry2Ae protein is expressed in the ppm levels (see Allergenicity Assessment above, and Tables B.1 and B.4 in this Appendix). These levels are several orders of magnitude lower than the amounts of Cry2Ae protein shown to have no toxicity. Even if low levels of dietary exposure should occur, no harm is expected, based upon the lack of mammalian toxicity and the rapid digestibility demonstrated for the Cry2Ae protein. 




c. Cumulative Effects from Substances with a Common Mechanism of Toxicity

Section 408(b)(2)(D)(v) of FFDCA requires that, when considering whether to establish, modify, or revoke a tolerance, the Agency consider "available information" concerning the cumulative effects of a particular pesticide's residues and "other substances that have a common mechanism of toxicity."

Cry2Ae protein is not considered to be toxic, and does not share a common mechanism of toxicity with any other substance. In addition, Cry2Ae protein does not appear to produce a toxic metabolite produced by other substances. EPA concludes that there are no cumulative effects associated with Cry2Ae protein that need be considered. For information regarding EPA's efforts to determine which chemicals have a common mechanism of toxicity and to evaluate the cumulative effects of such chemicals, see EPA's website at http://www.epa.gov/pesticides/cumulative 

d. Determination of Safety for U.S. Population, Infants and Children 

The data submitted and cited regarding potential health effects for the Cry2Ae protein include the characterization of the expressed Cry2Ae protein in cotton, as well as the acute oral toxicity, heat stability, and in vitro digestibility of the protein. The results of these studies were used to evaluate human risk, and the validity, completeness, and reliability of the available data from the studies were also considered. 
As discussed more fully in Appendix B.II.a., above, the acute oral toxicity data submitted supports the prediction that the Cry2Ae protein would be nontoxic to humans. Moreover, Cry2Ae protein has no sequence similarity to any known toxin. Because of this lack of demonstrated mammalian toxicity, no protein residue chemistry data for Cry2Ae were required for a human health effects assessment. Even so, expression level analysis showed Cry2Ae protein is present at relatively low levels. Dietary exposure is expected to be correspondingly low. 
Since Cry2Ae is a protein, its allergenic potential was considered as part of the toxicity assessment. Data considered as part of the allergenicity assessment include that the Cry2Ae protein: 
   a) came from Bacillus thuringiensis, which is not a known allergenic source; 
   b) showed no sequence similarity to known allergens; 
   c) was readily degraded by pepsin; and 
   d) was not glycosylated when expressed in the plant. 
On this basis, EPA concluded there is a reasonable certainty that Cry2Ae protein will not be a food allergen. Considered together, the lack of mammalian toxicity at high levels of exposure to the Cry2Ae protein and the minimal potential for that protein to be a food allergen demonstrate the safety of the product at levels well above possible maximum exposure levels anticipated in the crop.

Finally, and specifically with regards to infants and children, FFDCA section 408(b)(2)(C) provides that EPA shall assess the available information about consumption patterns among infants and children, special susceptibility of infants and children to pesticide chemical residues, and the cumulative effects on infants and children of the residues and other substances with a common mechanism of toxicity. In addition, FFDCA section 408(b)(2)(C) provides that EPA shall apply an additional tenfold margin of safety for infants and children in the case of threshold effects to account for prenatal and postnatal toxicity and the completeness of the data base unless EPA determines that a different margin of safety will be safe for infants and children. 

Based on all of the available information described in this BRAD,, the Agency concludes that there is a reasonable certainty that no harm will result to the U.S. population, including infants and children, from aggregate exposure to residues of the Cry2Ae protein and the genetic material necessary for its production in cotton. This includes all anticipated dietary exposures and all other exposures for which there is reliable information. The Agency also concluded that there are no threshold effects of concern and, as a result, that an additional margin of safety for infants and children is unnecessary in this instance.

e. Other Considerations 

1) Endocrine Disruptors 

The pesticidal active ingredient is a protein that breaks down readily in dietary exposure and is derived from a source not known to exert an influence on the endocrine system. Therefore, the Agency is not requiring information on the endocrine effects of the plant-incorporated protectant at this time.

2) Analytical Method 

BCS provided a validation study for EnviroLogix's QuantiPlate(TM) Kit for Cry2A (Cat. No. AP005), which uses enzyme-linked immunosorbent assay (ELISA) analyses for the detection of Cry2Ae protein residues in Event GHB119 cotton. Results showed the test kit is able to detect Cry2Ae protein residues in cotton with sufficient accuracy, precision, and sensitivity. However, a validation study conducted by an independent 3[rd] party laboratory is still needed to evaluate the performance of the ELISA test kit as the designated residue analytical method for the detection of Cry2Ae protein expressed in Event GHB119 cotton.

3) International Residue Limits

In making its tolerance decisions, EPA seeks to harmonize U.S. tolerances with international standards whenever possible, consistent with U.S. food safety standards and agricultural practices. In this context, EPA considers the international maximum residue limits (MRLs) established by the Codex Alimentarius Commission (Codex), as required by FFDCA section 408(b)(4). The Codex Alimentarius is a joint U.N. Food and Agriculture Organization/World Health Organization food standards program, and it is recognized as an international food safety standards-setting organization in trade agreements to which the United States is a party. EPA may establish a tolerance that is different from a Codex MRL; however, FFDCA section 408(b)(4) requires that EPA explain the reasons for departing from the Codex level. No Codex maximum residue level exists for the plant-incorporated protectant Bacillus thuringiensis Cry2Ae protein. 

f. Overall Safety Conclusion

There is a reasonable certainty that no harm will result from aggregate exposure to the U.S. population, including infants and children, to the residues of Cry2Ae protein and the genetic material necessary for its production. This includes all anticipated dietary exposures and all other exposures for which there is reliable information. The Agency has arrived at this conclusion because, as discussed above, no toxicity to mammals has been observed, nor any indication of allergenicity potential for the plant-incorporated protectant. 
The data submitted and reviewed for Cry2Ae protein support the petition for an exemption from the requirement of tolerance for Bacillus thuringiensis Cry2Ae protein when used as plant -  incorporated protectant in or on cotton food/feed commodities. 
G. Supporting Data 

The human health studies submitted to support the safety of Cry2Ae protein are summarized in Table B.3 below. For detailed information of each study, please refer to the individually reviewed studies in the Data Evaluation Reports (DERs) according to its respective assigned MRID number).   

Table B.3. Human Health Data for Event GHB119 cotton expressing Bt Cry2Ae protein
                          (EPA File Symbol 264-RNOL)

                                  Study Title
                 B.3. Summary  -  Human health  -  Cry2Ae PIP
                                    MRID No
Description of the Amino Acid Sequence of the Cry2Ae Protein
The 631 amino acid sequence of the Cry2Ae protein is provided in this submission. No information is provided on how the sequence was determined, though it was presumably deduced from the inserted sequence of the gene, and no information is provided on whether the sequence was verified by mass spectrometry or N-terminal amino acid sequencing. SDS-PAGE and western blot analyses of the plant-produced protein are provided in MRID 46708907, which indicate that the protein has the expected molecular weight. 
Classification: SUPPLEMENTAL; Preliminary study used to support EUP
                                   46708902
Cry2Ae (GEM2) Protein: Overall Amino Acid Sequence Homology Search with Known Toxins and Allergens[4]
The amino acid sequence of the Cry2Ae protein was compared to sequences contained in publicly available databases (SwissProt, trEMBL, GeneSeq-Prot, PIR, PDB, DAD, and GenPept.) using the BLASTP 2.2.2 algorithm and the BLOSUM62 scoring matrix. No similarities between the Cry2Ae protein and known allergens or toxins, other than Cry toxins, were found using the criterion of 35% identity over an 80 amino acid segment. 

Classification: SUPPLEMENTAL; Sufficient to support EUP and tolerance exemption; Data gaps addressed for Sec. 3 registration in MRID No. 47641910
                                   46708903
Cry2Ae (GEM2) Protein: Epitope Homology and N-Glycosylation Searches[4]
The amino acid sequence of the Cry2Ae protein was compared to known protein sequences in the Allergen database to identify any similarities between short segments of amino acids. Searches were also performed to identify any possible N-glycosylation motifs. No significant similarities between Cry2Ae protein and known allergens based on a 100% identity over an 8 amino acid segment were found. Eleven potential N-glycosylation sites were found in the Cry2Ae protein sequence. 

Classification: ACCEPTABLE 
                                   46708904
Cry2Ae (GEM2) Protein in vitro Digestibility Study in Simulated Gastric Fluid[4]
The digestibility of the trypsin resistant core of microbially produced Cry2Ae protein was investigated in simulated gastric fluid containing pepsin. Horseradish peroxidase and ovalbumin were included in the study as controls. The trypsinized Cry2Ae protein was degraded within two minutes of incubation in simulated gastric fluid at pH 1.2. Horseradish peroxidase and ovalbumin were degraded rapidly and slowly, respectively, as expected. However, because the protein that is produced in cotton is the full-length Cry2Ae protein, digestibility of the full-length protein is more relevant for the risk assessment.
Classification: SUPPLEMENTAL- Data gaps were addressed for EUP and Sec. 3 registration in MRID No. 47125102 
                                   46708905
                                       
                                       
Cry2Ae (GEM2) Protein - Acute Toxicity by Intravenous Injection in the Mouse[4]
The acute toxicity of the trypsin resistant core of microbially produced Cry2Ae was investigated by intravenous injection in mice. The protein (>95% purity) was suspended in PBS buffer (phosphate buffered saline) and administered as a single 1 or 5 mg/kg body weight intravenous dose. Melittin protein was used as a positive control, and aprotinin protein and buffer alone were used as negative controls. Mice were observed for clinical signs daily for the duration of the study (15 days), and gross necropsy was performed at the end of the study. Four out of five mice treated with melittin at a dose of 5 mg/kg died on day 1. No other deaths occurred during the study. No treatment-related adverse effects were observed in the mice administered Cry2Ae protein at either dose. 
Classification: SUPPLEMENTAL; Data gaps were addressed for EUP in MRID No. 47076902  and no additional data were required 
                                   46708906
Analysis to Determine if the Cry2Ae (GEM2) Protein from Cotton Leaves is Glycosylated[4]
Cry2Ae protein, purified from Bacillus thuringiensis and from cotton leaves from Events GHB119, GHB714, and GHB866, was analyzed for glycosylation using a commercially available glycosylation test kit. The results indicate that the Cry2Ae protein is not glycosylated. 
Classification: ACCEPTABLE 
                                   48471901
                   [This study supersedes MRID No. 46708907]
Cry2Ae (GEM2) Protein: Acute Toxicity by Oral Gavage in Mice[4]
The acute oral toxicity of Cry2Ae protein was assessed by administering groups of five female mice Bacillus thuringiensis-produced Cry2Ae protein by oral gavage at a dose of 2000 mg/kg body weight. There were no mortalities, and no treatment-related adverse effects observed. Therefore, the acute oral LD50 of the Cry2Ae protein is greater than 2000 mg/kg body weight. 
Classification: ACCEPTABLE 
                                   47076902
Cry2Ae (GEM2) Protein: In Vitro Digestibility Study in Simulated Gastric Fluid[4]
The susceptibility of the Cry2Ae protein to proteolytic degradation in simulated gastric fluid containing pepsin (SGF) was investigated using Bacillus thuringiensis-produced full-length Cry2Ae protein. The band for the Cry2Ae protein was no longer visible on the Coomassie blue stained gel after incubation in SGF for 0.5 minutes. In addition, smaller degradation fragments were not observed. Horseradish peroxidase and ovalbumin were degraded rapidly and slowly, respectively, as expected. This study indicates that the Cry2Ae protein is rapidly digested in SGF at pH 1.2. 
Classification: ACCEPTABLE 
                                   47125102
    [This study supersedes MRID No. 46708905 and confirms MRID No. 47125102]
Cry2Ae Protein: Heat stability study
An in vitro heat stability study was conducted on bacterial-expressed Cry2Ae protein at temperatures of 60, 75 and  90°C for incubation periods of 0, 10, 30 and 60 minutes and examined by SDS-PAGE and by Western blot analysis. Results of the SDS-PAGE showed a ~59.3 kDa band for Cry2Ae protein and no significant deterioration after incu -  - ba - tion at 60ºC at all time points, but partial degradation at 75ºC for 10 through 60 minutes. After 30 minutes at 90ºC, the band was visible, but markedly decreased and by 60 minutes was no longer present. Results of the Western blot analyses showed a ~59.3 kDa immunoreactive band for Cry2Ae protein and no signifi - cant deterioration after incu - bation at 60ºC for 10, 30, or 60 minutes. At incubation of 75ºC, the protein was partially stable for 10 minutes and further decreased in stain intensity through 60 minutes. At incubation of 90ºC, the band was visible at 10 minutes, but no longer present after 30 through 60 minutes. Results showed consistent patterns of decreasing structural stability of the Cry2Ae protein in between both analyses tested. This study indicates that the Cry2Ae protein is heat labile at 90°C at 60 minutes.
Classification: ACCEPTABLE
                                   47641911
Toxicology (Human Health Assessment) Summary of Analysis of Open Reading Frame Homology Searches
The purpose of this study was to evaluate newly created ORFs that span the 5' or 3' junction of the transgenic insert (T-DNA) in cotton Event GHB119 and determine if any newly created chimeric ORFs could give rise to amino acid sequences with toxic or allergenic properties. The ORF search was performed by means of the GetORF search program from the EMBOSS (European Molecular Biology Open Software Suite). BLASTP or FindPatterns algorithms were used to search publically available databases (Uniprot-Swissprot, Uniprot-TrEMB, DAD, PDB, and GenPept). Three ORFs were found, one spanning the 5' junction and two spanning the 3' junction. No significant similarities were identified between any of the putative amino acid sequences generated by these ORFs and sequences found in databases, including known allergens and toxins.
Classification: ACCEPTABLE
                                   47641906
Cry2Ae Protein: In vitro digestibility study in simulated intestinal fluid
An in vitro digestibility study was conducted on bacterial-expressed Cry2Ae protein in simu - lated human intestinal fluid (SIF) containing pancreatin (pH 7.5) to assess the susceptibility of Cry2Ae protein to proteolytic degradation. The Cry2Ae protein was degraded into several stable fragments in the presence of pancreatin, at pH 7.5 during the 60 minute digestion. 
Classification: ACCEPTABLE
                                   47641907
                                       
Cry2Ae protein with transit peptide epitope homology and N-glycosylation searches

No significant amino acid sequence similarities between eight linear contiguous amino acid blocks of the Cry2Ae protein with known aller - gens in the publically available AllergenOnline database (release 8.0, 1313 sequences) were found. Results suggest the Cry2Ae protein should not cross-react with known allergenic epitopes. Eleven potential N-glycosylation sites were identified.
Classification: ACCEPTABLE
                                   47641908
                   [This study supersedes MRID No. 47076904]
Cry2Ae protein with transit peptide overall amino acid sequence homology search with known toxins and allergens
A. Bioinformatic searches were conducted to evaluate any significant amino acid sequence similarities between the Cry2Ae protein and any known toxins in public available databases (Uniprot_SwissProt, Uniprot_TrEMBL, PDB, DAD, and GenPept.) using the FASTA algorithm and the BLOSUM62 scoring matrix; and any known allergens in the AllergenOnline database (release 8.0) using the FASTA algorithm and  BLOSUM50 scoring matrix. Comparisons were conducted on the complete amino acid sequence of the test protein with matches of >35% identity over an 80 amino acid sequence segment; and then sequentially compared using sub-divided sequences of the test protein into 80 amino acid segments with all known allergens present in the database. No similarities between the Cry2Ae protein and known allergens or toxins, other than Cry toxins, were found.
Classification: ACCEPTABLE
                                   47641909
                   [This study supersedes MRID No. 47076903]
Update on Cry2Ae protein: In vitro digestibility study in human simulated gastric fluid, overall amino acid sequence homology search with known toxins and allergens, and epitope homology and N-glycosylation searches 
Bioinformatic studies on Cry2Ae protein were repeated in 2008 and showed no amino acid sequence similarities between Cry2Ae protein and known toxins and allergens (MRID No. 47641909); and no amino acid sequence similarities with known allergen epitopes and N-glyco - sylation sites (MRID No. 47641908), thus confirming original results from previously conducted studies done in 2005 (MRIDs No. 46708903 and MRID 46708904, respectively). An in vitro digestibility study of Cry2Ae protein in human simulated gastric fluid containing pepsin was repeated in 2007 (MRID No. 47125102) and showed rapid degradation of Cry2Ae protein to pepsin, thus confirming original results from the previously conducted in vitro digestibility SGF study in 2005 (MRID No. 46708905).
Classification: SUPPLEMENTAL- This study is a supplemental  to MRIDs 46708905, 46708903 and 46708904
                                   47641910
[This study is a supplemental report to MRIDs 46708905, 46708903 and 46708904]


B.III. Product characterization and human health effects for TwinLink(TM) (T304-40 x GHB119) Combination PIP Cotton Product (OECD Unique Identifier: BCS-GH4-7 x BCS-GH5-8) Expressing Cry1Ab, PAT, and Cry2Ae Proteins

BCS submitted data in support of FIFRA Sec. 3 registration of TwinLink(TM) cotton (EPA File Symbol 264-RNOA) for use as an alternative insect resistant and herbicide tolerant combination PIP cotton product. TwinLink(TM) cotton was developed by conventional breeding of the parental Events T304-40 x GHB119 to express Cry1Ab, Cry2Ae and PAT proteins. Event T304-40 cotton encodes the cry1Ab gene derived from Bt subsp. berliner (described in Höfte et al. 1986) to express Cry1Ab insecticidal protein, which also confers resistance to lepidopteran pest species. The safety of Cry1Ab protein, as expressed in the cotton Event T304-40, was previously reviewed by the Agency and the data were found acceptable to support registration of Event T304-40 cotton (US EPA 2006, 2008, and 2011a and c). A tolerance exemption exists for Cry1Ab protein residues when used as plant-incorporated protectant in all food commodities (40 CFR § 174.511) and all final data submitted for its registration were found acceptable (US EPA 2011c). Considerable safety data have been established for both Cry2Ae protein expressed in GHB119 and Cry1Ab protein expressed in T304-40 cotton. The applicant submitted the following equivalency studies in order to bridge from these existing acceptable toxicological data for the parental Events for registration of TwinLink(TM) Cotton.

a. Product Characterization

TwinLink(TM) cotton (T304-40 x GHB119 combination PIP product) contains cry1Ab, cry2Ae, and bar transgenes. Events T304-40 and GHB119 cotton were individually developed by Agrobacterium-mediated transformation to insert T-DNA containing the cry1Ab and cry2Ae gene cassettes via plasmid vectors pTDL008 and pTEM12, respectively. The resulting cotton lines were then crossed by conventional breeding to create TwinLink(TM) (GHB119 x T304-40) cotton.

Characterization of DNA isolated from TwinLink(TM) (T304-40 x GHB119) cotton using restriction enzyme digests and Southern blot analyses were used to confirm the genetic stability and integrity of the T304-40 and GHB119 transgenic inserts in TwinLink(TM) cotton. The analyses showed the presence of the expected hybridization fragments in all tested transgenic DNA samples when hybridized with each of the three probes. In addition, results showed the cry1Ab, cry2Ae, and bar DNA inserts are stably integrated in TwinLink(TM) cotton, based on Southern blot analyses of combination PIP product plants obtained after five generations from the original TwinLink(TM) cross. Data from the comparative Southern blot analyses confirmed the integrity and genetic stability of the transgenic cry1Ab, cry2Ae, and bar inserts from the individual parental events as preserved during conventional breeding to produce TwinLink(TM) cotton.

b. Analytical Detection Methods

BCS provided validation studies for commercial enzyme-linked immunosorbent assay (ELISA) test kits for the analytical detection of Cry2Ae and Cry1Ab protein residues in cottonseed and cotton leaf tissue. The studies showed the test kits are able to detect Cry2Ae and Cry1Ab protein in Events GHB119 and T304-40 cotton, respectively, with sufficient accuracy, precision, and sensitivity. Verification of the specific ELISA test kit method to be used as the proposed residue analytical method for the detection of Cry1Ab protein in Event T304-40 cotton is needed. In addition, individual validation studies conducted by an independent third party laboratory are still needed to evaluate the performance of each ELISA test kit as the designated residue analytical method for the detection of Cry2Ae and Cry1Ab proteins expressed in cotton Events GHB119 and T304-40, respectively.

c. Protein Expression

Protein expression levels were provided for Cry2Ae, PAT, and Cry1Ab in different plant tissues from T304-40 x GHB119 cotton tested at key developmental growth stages at 4 field sites. The average protein expression levels are shown below in Table 4. The Cry2Ae and Cry1Ab protein levels in TwinLink(TM) cotton are comparable to those observed in plant tissues from its parental Events GHB119 and T304-40 cotton. The PAT protein expression levels in TwinLink(TM) cotton were higher than those expressed in its parental events. These results were expected since TwinLink(TM) cotton inherited two copies of the bar gene, which were individually inserted in its parental events.

                                       
Table B.4. Mean Expression Levels of Cry2Ae, PAT and Cry1Ab Proteins from TwinLink(TM) cotton

                                  Tissue Type
                                    Cry2Ae
                    (ug/g dry weight + standard deviation)
                                      PAT
                    (ug/g dry weight + standard deviation)
                                    Cry1Ab
                    (ug/g dry weight + standard deviation)
                                Leaves (T1-T3)*
                            18 +- 6.4  -  70  +- 35
                           250 +-120  -  1200 +- 650
                           3.0 +- 2.1  -  8.8 +- 3.8
                                 Squares (T3)
                                   12 +- 4.7
                                  370 +- 160
                                  8.2 +- 3.7
                                  Bolls (T3)
                                   4.7 +-1.7
                                  240 +- 100
                                  6.5 +- 2.9
                                Cottonseed (T4)
                                  6.7 +- 3.3
                                  280 +- 130
                                   13 +- 6.5
                                Roots (T1-T3)*
                           3.9 +- 1.9  -  11 +- 5.3
                           170 +- 80  -  290 +- 130
                           2.5 +- 1.4  -  3.8 +- 1.9
      *Ranges reflect means tested at different growth stages 
	 

d. Conclusion

The DNA characterization study and protein expression analyses field study considered with comparable biological activity in between the parental cotton Events GHB119 and T304-40 and no evidence of either a synergistic or antagonistic interaction between Cry2Ae and Cry1Ab proteins in tobacco budworm (TBW, Heliothis virescens) (see Dean, B. et al. 2010; MRID No. 48480009) reviewed in the ecological risk assessment memo for this product (US EPA 2011b) demonstrate that the Cry2Ae and Cry1Ab proteins expressed in the parental cotton Events GHB119 and T304-40 are biochemically and functionally equivalent to the Cry2Ae and Cry1Ab proteins expressed in TwinLink(TM) cotton. Thus, the existing toxicological data of parental Events GHB119 and T304-40 cotton can be bridged to support the safety of TwinLink(TM) Cotton. 

e. Supporting Data

The product characterization studies submitted to support the biochemical equivalence of Cry2Ae and Cry1Ab proteins expressed in TwinLink(TM) cotton to Cry2Ae and Cry1Ab expressed in its constituent parental Events GHB119 and T304-40, respectively, are summarized in Table 5 below. For detailed information of each study, please refer to the individually reviewed studies in the Data Evaluation Reports (DERs) according to its respective assigned MRID number).   






   Table B.5. Product Characterization Data for Combination PIP TwinLink(TM)
                          (EPA File Symbol 264-RNOA)

                                  Study Title
             B.5. Summary - TwinLink(TM) - Product Characterization
                                   MRID No.
TwinLink(TM)  Product Characterization - Molecular Characterization
Southern blot analyses were used to characterize the genomic DNA to confirm the integrity of the GHB119 and T304-40 inserts in the combination PIP product T304-40 x GHB119. Samples from T304-40 x GHB119 cotton showed band patterns obtained using each probe were repeatable as those observed for the individual events, indicating that no DNA rearrangements occurred during the production of TwinLink(TM) developed by conventional breeding of the parental Events T304-40 x GHB119.

However, no Southern blot analyses were conducted using the pat/bar as a probe on the parental events and TwinLink(TM) Cotton. In addition, details were missing on the test materials, methods used to conduct the study. There were also no plasmid maps or schematic depictions  included to show the location of the cleavage sites for each restriction enzyme, the probe positions, and the predicted sizes of each DNA fragments after digestion. Lastly, the lanes for the negative control and the positive control appear to be mislabeled, since the negative and positive control results were opposite to those expected in both gel figures.  

CLASSIFICATION:  SUPPLEMENTAL- Data gaps were addressed in MRID No. 47640007
                                  47640001  
Protein Expression Analysis for T304-40, GHB119 and TwinLink(TM) Transgenic Cotton[5]
This study was conducted to characterize and compare the levels of protein expression for Cry1Ab, Cry2Ae, and PAT expressed in Events T304-40, GHB119,  and the combination PIP TwinLink(TM) cotton, developed by conventional breeding of the parental Events T304-40 x GHB119. The trial contained 50 plants for each cotton line grown under typical greenhouse conditions including the application of insecticides and fertilizer. Samples of leaves, squares and grain were collected from five randomly selected cotton plants from each genotype. The Cry1Ab, Cry2Ae and PAT proteins were extracted from the plant tissue matrices to measure the amount of protein expressed in the cotton lines, using commercially available enzyme-linked immunosorbent assay (ELISA) test kits. There was considerable variation in the levels of protein expression reported for each PIP protein expressed in the parental events compared to the levels of protein expression in TwinLink(TM) Cotton; as well as  significant differences reported within the ranges of protein expression levels for certain tissue types, although not statistically analyzed.  

However, the significant differences may be attributed to the limited number of samples, replicates, and trials since this study was conducted as a single trial in a greenhouse. Therefore, the dataset used to characterize the protein expression levels may not be representative of the protein levels expressed in the field. In addition, the study does not account for variability in the levels of protein expressed under field conditions, where different environmental factors may also have an effect.  

CLASSIFICATION:  SUPPLEMENTAL- Data gaps were addressed in MRID No. 47640008  
                                 47640003     
Analyses of Raw Agricultural Commodity (Fuzzy Seed) of TwinLink(TM) Cotton for PAT/bar, Cry2Ae and Cry1Ab and its Non-transgenic Counterpart for PAT/bar, Cry2Ae and Cry1Ab Proteins
The purpose of this study was to quantify the expression of Cry1Ab, Cry2Ae and PAT proteins in the fuzzy seeds of transgenic TwinLink(TM) cotton and the non-transgenic Coker 315 cotton line using enzyme-linked immunoabsorbent assay (ELISA). Fuzzy seed were separated into lint coats and kernels and these fractions were analyzed separately for Cry1Ab, Cry2Ae and PAT proteins and for total extractable protein. The respective values for the kernel and lint coat fractions were combined on a weight basis to calculate the respective values for the fuzzy seed that was received from the field. The limit of detection (LOD) and limits of quantitation (LOQ), the extract dilution factors, extraction efficiencies, standard curves (ng/mL), and the calculations for the protein extractions from each tissue samples were reported to establish the validity (sufficient sensitivity and extraction efficiencies) of the ELISA method used in this study. 

The average Cry1Ab protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, on a dry weight basis, was 1.32 +- 0.517 ug/g and 1.57 +- 0.317ug/g, respectively. As a percent of total crude protein, Cry1Ab protein comprised an average of 0.000577 +- 0.000204% and 0.000707 +- 0.000155% of the total crude protein in unsprayed and sprayed fuzzy seed, respectively. 

The average Cry2Ae protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, on a dry weight basis, was 9.35 +- 2.93ug/g and 10.7 +- 4.17 ug/g, respectively. As a percent of total crude protein, Cry2Ae protein comprised an average of 0.00420 +- 0.00129 % and 0.00485 +- 0.00204 % of the total crude protein in unsprayed and sprayed fuzzy seed, respectively. 
The average PAT protein content of transgenic fuzzy seed from all test sites, without and with herbicide spraying, on a dry weight basis, was 162 +- 32.5 ug/g and 163 +- 27.4 ug/g, respectively. As a percent of total crude protein, PAT protein comprised an average of 0.0726 +- 0.0121% and 0.0737 +- 0.0151% of the total crude protein in unsprayed and sprayed fuzzy seed, respectively.

CLASSIFICATION:  ACCEPTABLE
                                   47640004
Structural stability analysis of Gossypium hirsutum combined Event TwinLink(TM) x GlyTol (GHB614)
The purpose of this study was to confirm the integrity and genetic stability of the intended transgenic inserts contained in combination PIP TwinLink(TM) cotton plants developed by conventional breeding of the parental cotton Events T304-40 (expressing Cry1Ab and PAT proteins) and GHB119 (expressing Cry2Ae and PAT proteins). TwinLink(TM) cotton contains the cry1Ab and bar (cry1Ab-bar) genes on one DNA insert from T304-40 cotton and the cry2Ae and bar (cry2Ae-bar) genes on a second DNA insert from GHB119 cotton. Genomic DNA was prepared from 24 individual combination PIP product cotton plants and was subjected to Southern blot analysis using the cry1Ab-bar TDNA probe, the cry1Ab gene probe, and the cry2Ae-bar TDNA probe. The analyses showed the presence of the expected hybridization fragments in all tested transgenic DNA samples when tested with each of the three probes. In addition, results showed the cry1Ab, cry2Ae, and bar DNA inserts are stably integrated in TwinLink(TM) cotton, based on Southern blot analyses of combination PIP product plants obtained after five generations from the original TwinLink(TM) cross. Therefore, data from the comparative Southern blot analyses confirmed the integrity and genetic stability of the transgenic cry1Ab, cry2Ae, and bar inserts from the individual parental events as preserved during conventional breeding to produce TwinLink(TM) Cotton.

CLASSIFICATION:  ACCEPTABLE
                                   48480007
                   [This study supersedes MRID No. 47640001]
Protein expression analysis for GlyTol x TwinLink(TM) (GHB614 x T304-40 x GHB119) cotton grown in the field, USA, 2010
The purpose of this study was to use enzyme-linked immunosorbent assay (ELISA) methods to determine the expression of CrylAb, Cry2Ae, and PAT/bar proteins in the combination PIP product: GlyTol x TwinLink(TM) (GHB614 x T304-40 x GHB119). Cotton Event T304-40 expresses the Cry1Ab and PAT/bar proteins, Event GHB119 expresses the Cry2Ae and PAT/bar proteins, and Event GHB614[1] expresses the 2mEPSPS (5-enolpyruvyl-shikimate-3-phosphate synthase) protein. Combination PIP product cotton tissue samples from leaves, roots, squares, bolls, and fuzzy seeds were collected from three different field trial sites at one or more developmental stages (T1 though T4). A total of 36 samples of each tissue type (matrix) were assayed for protein concentrations. Expression of all proteins (CrylAb, Cry2Ae and PAT/bar) was observed in all matrices (leaves (T1, T2, and T3), roots (T1, T2, and T3), squares (T3), bolls (T3), and fuzzy seed (T4)). 

Detailed information was provided on the limit of detection (LOD) and limits of quantitation (LOQ), the extract dilution factors, extraction efficiencies, standard curves (ng/mL), and the calculations for the protein extractions from each tissue samples for the purposes of validating the ELISA methods used in this study. Overall, the levels of protein expression for Cry2Ae, Cry1Ab, and PAT proteins in the combination PIP maize product were comparable to the protein expression levels in its respective parental cotton Events GHB119 and T304-40. These results can also be used to establish the biochemical equivalence of the Cry2Ae, Cry1Ab and PAT proteins expressed in combination PIP relative to the PIP parental Events GHB119 and T304-40, when these parental lines are combined via traditional breeding methods.

CLASSIFICATION:  ACCEPTABLE  
                                   48480008
                   [This study supersedes MRID No. 47871604]
REFERENCES  -  Cry2Ae and TwinLink(TM) Cotton PIPs
     Bernstein, I.L., Bernstein, J.A., Miller, M., Tierzieva, S., Bernstein, D.I., Lummus, Z., Selgrade, M.K., Doerfler DL, Seligy VL. (1999) Immune responses in farm workers after exposure to Bacillus thuringiensis pesticides. Environ Health Perspect. 107(7):575-82
     
CAC (2003) Alinorm 03/34: Joint FAO/WHO Food Standard Programme. Codex Alimentarius Commission, Twenty-Fifth Session, 30 July 2003. Rome, Italy. Appendix III: Guideline for Conduct of Food Safety Assessments of Foods Derived from Recombinant-DNA Plants; Appendix IV: Annex on Assessment of Possible Allergenicity. CAC, 47 - 60
Dean, B., Jansen., S., Becker, A. (2010) Insect bioassay of plant material expressing cotton Events T304-40 GHB119 and TwinLink(TM) (T304-40 x GHB119). Bayer CropScience Internal Report -Insect Bioassay from Monheim. MRID No. 48480009 
Höfte H., de Greve H., Seurinck J., Jansens S., Mahillon J., Ampe C., Vandekerckhove J., 
      Vanderbruggen H., van Montagu M., Zabeau M.,Vaeck M. (1986) Structural and functional analysis of a cloned delta endotoxin of Bacillus thuringiensis berliner 1715. Eur. J. Biochem. 161:273-280
Sjoblad, Roy D. et al. (1992) Toxicological Considerations for Protein Components of Biological Pesticide Products" Regulatory Toxicology and Pharmacology 15: 3-9 
Thompson C., Van Montagu M. and Leemans J. (1987) Engineering herbicide resistance in plants by expression of a detoxifying enzyme. EMBO Journal. 6(9): 2513-2518
U.S. EPA. 2006. "The product characterization and human health data submitted in support of the EUP."  Memorandum from R. Edelstein, through J. Kough, to S. Matten; dated January 27, 2006. US Environmental Protection Agency. Washington DC
U.S. EPA. 2008. Review of Product Characterization and Human health Data for an EUP Application for Bt. Cry2Ae Insecticidal Protein and Cry2Ae x Cry1Ab as Expressed in Cotton. Memorandum from R. Edelstein, through J. Kough, to S. Bacchus; dated February 12, 2008. US EPA. Washington DC
U.S. EPA. 2009. Residue Chemistry Test Guidelines: OPPTS 860.1340 Residue Analytical Method (EPA 712 - C - 96 - 174); Document ID: EPA-HQ-OPPT-2009-0155-0007; Docket ID: EPA-HQ-OPPT-2009-0155 
U.S. EPA. 2011a. Review of Product Characterization Data in support of Sec. 3 Registration of Plant-Incorporated Protectant (PIP) cotton Events: GHB119 cotton expressing Bacillus thuringiensis Cry2Ae protein (EPA File Symbol 264-RNOU), T304-40 cotton expressing Bt CrylAb (EPA File Symbol 264-RNOL); and TwinLink(TM) cotton (EPA File Symbol 264-RNOA), its associated combination PIP product developed by conventional breeding of cotton Events GHB119 and T304-40."  Memorandum from A. Waggoner, through J. Kough, to S. Bacchus and D. Greenway; dated February 4, 2011. US EPA. Washington DC
U.S. EPA. 2011b. Environmental Risk Assessment for Bacillus thuringiensis (Bt) Cry1Ab and Bt Cry2Ae insect control proteins as expressed in Event T304-40 cotton and GHB119 cotton, respectively, and in the combined Event, T304-40 x GHB119 (TwinLink(TM)) cotton (EPA File Symbols 264-RNOU, -RNOL, -RNOA), submitted by Bayer CropScience LP. Memorandum from S. Borges to S. Bacchus; dated October 13, 2011. US EPA. Washington DC
US EPA. 2011c. Review of Product Characterization and Human Health Data in support of Sec. 3 Registration of Plant-Incorporated Protectant (PIP) Event T304-40 cotton expressing Bacillus thuringiensis Cry1Ab insecticidal protein (EPA File Symbol 264-RNOU). Memorandum from A. Waggoner, through J. Kough, to D. Greenway; dated November 30, 2011. US EPA. Washington DC
U.S. FDA. 1992. Statement of Policy: Foods Derived From New Plant Varieties. Federal Register: Vol. 57 No. 104 p. 22984. Dated: of May 29, 1992. Department Of Health And Human Services- Food and Drug Administration (Docket No. 92N-0139) http://www.cfsan.fda.gov/~lrd/biotechm.html#reg 
Van Rie J. 2004. Comparative analysis of the Cry1Ab protein amino acid sequence. Project Number: None; Unpublished study prepared by Bayer CropScience LP. MRID No. 46474406.
APPENDIX C: Ecological and environmental effects  -  Bacillus thuringiensis Cry1Ab, Cry2Ae and TwinLink(TM) Cotton plant incorporated protectant

 BCS applied for a FIFRA Section 3 registration for two plant incorporated protectants (PIPs). These include:
   * Cry1Ab protein and the genetic material necessary for its production in Event T304-40 transgenic cotton [Gossypium hirsutum] plant line;
   * Cry2Ae protein and the genetic material necessary for its production in transgenic Event GHB119 cotton plant line; and 
   * the associated pyramid in Event T304-40 x GHB119 (referred to hereafter as TwinLink(TM)) Cotton, which produces both proteins. 
 BPPD does not anticipate any adverse effects of the Cry1Ab and Cry2Ae proteins on nontarget organisms or on their abundance in any field population. BPPD additionally concludes that the proposed Section 3 registrations of Cry1Ab and Cry2Ae proteins as expressed in Event T304-40 x Event GHB119, or TwinLink(TM) Cotton will not result in direct or indirect effects to nontarget organisms, including federally-listed threatened or endangered species.  
 
 This assessment references data that were reviewed and found acceptable to demonstrate functional equivalence between the bacteria-derived Cry1Ab and Cry2Ae proteins used in toxicity testing and plant-expressed Cry1Ab and Cry2Ae proteins. The ecological risk assessment for the combined Events expressed in TwinLink(TM) cotton is also dependent on equivalent expression levels of Cry1Ab and Cry2Ae protein in TwinLink(TM) cotton plants compared to the T304-40 and GHB119 parent lines [see Product Characterization - Appendix A (Cry1Ab) and B (Cry2Ae)].
 

BACKGROUND			

BCS developed Event T304-40 and Event GHB119 transgenic cotton [Gossypium hirsutum] plant lines, which each express a Cry protein insect control gene. Event T304-40 expresses a Cry1Ab gene, which is derived from the soil bacterium, Bacillus thuringiensis (Bt) subsp. berliner, and Event GHB119 expresses a cry2Ae gene derived from Bt subsp. dakota. Both genes have been modified for their expression in plants. Cry1Ab and Cry2Ae proteins are intended to control several lepidopteran pests of cotton including cotton bollworm (Helicoverpa zea), pink bollworm (Pectinophora gossypiella), tobacco budworm (Heliothis virenscens) larvae, beet armyworm (Spodoptera exigua), and fall armyworm (Spodoptera frugiperda). Events T304-40 and GHB119 also express the PAT gene, which confers herbicide tolerance.

Experimental use permits (EUP) were previously granted by EPA to conduct field tests on Event T304-40, Event GHB119, and T304-40 x GHB119 cotton (EPA Reg. Nos. 264-EUP-140 and 264-EUP-143). BCS is currently requesting a FIFRA Sec. 3 registration for Bt insect control protein Cry1Ab as expressed in Event T304-40 cotton, Cry2Ae as expressed in Event GHB119 cotton, and the associated breeding stack, TwinLink(TM) cotton, produced by a traditional breeding cross. This memorandum contains the Agency's environmental risk assessment of the Cry1Ab and Cry2Ae proteins produced in T304-40 and GHB119 cotton lines, respectively, as well as in TwinLink(TM) Cotton. 
I.	Environmental Risk Assessment for Cry1Ab as Expressed in T304-40 Cotton and Cry2Ae as Expressed in GHB119 Cotton PIPs

The paragraphs below describe the process and rationale developed by BPPD for evaluating hazard of PIPs to nontarget organisms. This process is described in several of BPPD's documents and is presented again here as background information.

To minimize data requirements and avoid unnecessary tests, risk assessments are structured such that risk is determined first from estimates of hazard under "worst-case" exposure conditions. A lack of adverse effects under these conditions would provide enough confidence that there is no risk and no further data would be needed. Hence, such screening tests conducted early in an investigation tend to be broad in scope but relatively simple in design, and can be used to demonstrate acceptable risk under most conceivable conditions. When screening studies suggest potentially unacceptable risk additional studies are designed to assess risk under more realistic field exposure conditions. These later tests are more complex than earlier screening studies. Use of this "tiered" testing framework saves valuable time and resources by organizing the studies in a cohesive and coherent manner and eliminating unnecessary lines of investigation. Lower tier, high dose screening studies also allow tighter control over experimental variables and exposure conditions, resulting in a greater ability to produce statistically reliable results at relatively low cost.  

Tiered tests are designed to first represent unrealistic worst case scenarios and ONLY progress to real world field scenarios if the earlier tiered tests fail to indicate adequate certainty of acceptable risk. Screening (Tier I) non-target organism hazard tests are conducted at exposure concentrations several times higher than the highest concentrations expected to occur under realistic field exposure scenarios. This has allowed an endpoint of 50% mortality to be used as a trigger for additional higher-tier testing. Less than 50% mortality under these conditions of extreme exposure suggest that population effects are likely to be negligible given realistic field exposure scenarios. 

The EPA uses a tiered (Tiers I-IV) testing system to assess the toxicity of a PIP to representative non-target organisms that could be exposed to the toxin in the field environment. Tier I high dose studies reflect a screening approach to testing designed to maximize any toxic effects of the test substance on the test (non-target) organism. The screening tests evaluate single species in a laboratory setting with mortality as the end point. Tiers II  -  IV generally encompass definitive hazard level determinations, longer term greenhouse or field testing, and are implemented when unacceptable effects are seen at the Tier I screening level.

Testing methods which utilize the tiered approach were last published by the EPA as Harmonized OPPTS Testing Guidelines, Series 850 and 885 (EPA 712-C-96-280, February 1996). These guidelines, as defined in 40 CFR 152.20, apply to microbes and microbial toxins when used as pesticides, including those that are naturally occurring, and those that are strain-improved, either by natural selection or by deliberate genetic manipulation. Therefore, PIPs containing microbial toxins are also covered by these testing guidelines. 

The Tier I screening maximum hazard dose (MHD) approach to environmental hazard assessment is based on some factor (whenever possible >10) times the maximum amount of active ingredient expected to be available to terrestrial and aquatic non-target organisms in the environment (EEC). Tier I tests serve to identify potential hazards and are conducted in the laboratory at high dose levels which increase the statistical power to test the hypotheses. Elevated test doses, therefore, add certainty to the assessment, and such tests can be well standardized. The Guidelines call for initial screening testing of a single group or several groups of test animals at the maximum hazard dose level. The Guidelines call for testing of one treatment group of at least 30 animals or three groups of 10 test animals at the screening test concentration. The Guidelines further state that the duration of all Tier I tests should be approximately 30 days. Some test species, notably non-target insects, may be difficult to culture and the suggested test duration has been adjusted accordingly. Control and treated insects should be observed for at least 30 days, or, in cases where an insect species cannot be cultured, for 30 days, until negative control mortality rises above 20 percent. 

Failing the Tier I (10 X EEC) screening at the MHD dose does not necessarily indicate the presence of an unacceptable risk in the field but it triggers the need for additional testing. A less than 50% mortality effect at the MHD is taken to indicate minimal risk. However, greater than 50% mortality does not necessarily indicate the existence of unacceptable risk in the field, but it does trigger the need to collect additional dose-response information and a refinement of the exposure estimation before deciding if the risk is acceptable or unacceptable. Where potential hazards are detected in Tier I testing (i.e. mortality is greater than  50%), additional information at lower test doses is required which can serve to confirm whether any effect might still be detected at more realistic field [1X EEC] concentrations and routes of exposure. 

When screening tests indicate a need for additional data, the OPPTS Harmonized Guidelines call for testing at incrementally lower doses in order to establish a definitive LD50 and to quantify the hazard. In the definitive testing, the number of doses and test organisms evaluated must be sufficient to determine an LD50 value and, when necessary, the Lowest Observed Effect Concentration (LOEC), No Observed Adverse Effect Level (NOAEL) , or reproductive and behavioral effects such as feeding inhibition,  weight loss, etc. In the final analysis, a risk assessment is made by comparing the LOAEC to the EEC; when the EEC is lower than the LOAEC, a no risk conclusion is made. These tests offer greater environmental realism, but they may have lower statistical power. Appropriate statistical methods, and appropriate statistical power, must be employed to evaluate the data from the definitive tests. Higher levels of replication, the number of test species, and/or repetition are needed to enhance statistical power in these circumstances. 

Data that show less than 50 % mortality at the maximum hazard dosage level  -  (i.e. LC50, ED50, or LD50 >10 X EEC) is sufficient to evaluate adverse effects, making lower field exposure dose definitive testing unnecessary. It is also notable that the recommended >10X EEC maximum hazard dose level is a highly conservative factor. The published EPA Level of Concern [LOC] is 50% mortality at 5X EEC  (U.S. EPA 1998).  

Validation:  The tiered hazard assessment approach was developed for the EPA by the American Institute of Biological Sciences (AIBS) and confirmed in 1996 as an acceptable method of environmental hazard assessment by a FIFRA Scientific Advisory Panel (SAP) on microbial pesticides and microbial toxins. The December 9, 1999 SAP agreed that the Tiered approach was suitable for use with Plant-Incorporated Protectants (PIPs); however, this panel recommended that, for PIPs with insecticidal properties, additional testing of beneficial invertebrates closely related to target species and/or likely to be present in GM crop fields should be conducted. Testing of Bt Cry proteins on species not closely related to the target insect pest was not recommended, although it is still performed to fulfill the published EPA non-target species data requirements. In October 2000, another SAP also recommended that field testing should be used to evaluate population-level effects on non-target organisms. The August 2002 SAP, and some public comments, generally agreed with this approach, with the additional recommendation that indicator organisms should be selected on the basis of potential for field exposure to the subject protein (U.S. EPA 2000, 2001a, 2002, and 2004). 

Chronic studies: Since delayed adverse effects and/or accumulation of toxins through the food chain are not expected to result from exposure to proteins, protein toxins are not routinely tested for chronic effects on non-target organisms. However, the 30 day test duration requirement does amount to subchronic testing when performed at field exposure test doses. Proteins also do not bioaccumulate. The biological nature of proteins makes them readily susceptible to metabolic, microbial, and abiotic degradation once they are ingested or excreted into the environment. Although there are reports that some proteins (Cry proteins) bind to soil particles, it has also been shown that these proteins are degraded rapidly by soil microbial flora upon elution from soil particles.  

Conclusion: The tiered approach to test guidelines ensures, to the greatest extent possible, that the Agency requires the minimum amount of data needed to make scientifically sound regulatory decisions. The EPA believes that maximum hazard dose Tier I screening testing presents a reasonable approach for evaluating hazards related to the use of biological pesticides and for identifying negative results with a high degree of confidence. The Agency expects that Tier 1 testing for short-term hazard assessment will be sufficient for most studies submitted in support of PIP registrations. However, if long range adverse effects must be ascertained, then higher-tier longer-term field testing will be required. As noted above, the October 2000 SAP and the National Academy of Sciences (NAS 2000) recommended testing non-target organisms directly in the field. This approach, with an emphasis on testing invertebrates found in corn fields, was also recommended by the August 2002 SAP and was supported by several public comments. Based on these recommendations, the Agency has required field studies on long term invertebrate population/community and Cry protein accumulation in soils as a condition of registration due to the lack of baseline data on the potential for long-term environmental effects from the cultivation of PIP-producing plants.

Since the commercialization of Bt crops, the number of field studies published in scientific literature in combination with the post-registration field studies submitted to the Agency has accumulated to a level where empirical conclusions can be made. As a result, the issue of long range effects of cultivation of these Cry proteins on the invertebrate community structure in Bt crop fields has since been adequately addressed. Specifically, a meta-analysis of the data collected from 42 field studies indicated that non-target invertebrates are generally more abundant in Bt cotton and Bt maize fields than in non-transgenic fields managed with insecticides (Marvier et al., 2007). In addition, a comprehensive review of short and long term field studies on the effects of invertebrate populations in Bt corn and cotton fields indicated that no unreasonable adverse effects are taking place as a result of wide scale Bt crop cultivation (Sanvido et al. 2007). Another review of field tests published to date concluded that the large-scale studies in commercial Bt cotton have not revealed any unexpected non-target effects other than subtle shifts in the arthropod community caused by the effective control of the target pests (Romeis et al., 2006). Slight reductions in some invertebrate predator populations are an inevitable result of all pest management practices, which result in reductions in the abundance of the pests as prey.  

Overall, the Agency is in agreement with the conclusions of these studies and collectively, these results provide extensive data to support that Bt crops have not caused long term environmental effects on a population level to organisms not targeted by Bt proteins. Based on these considerations, regulatory testing of the specialist predators and parasitoids of target pests may eventually be considered unnecessary.   
 
      A. Environmental Exposure Assessment

Two separate SAP reports (October 2000 and August 2002) recommended that non-target testing of Bt Cry proteins should focus on invertebrate species exposed to the crop in which the protein(s) will be expressed. Following SAP recommendations, the EPA determined that non-target organisms with the greatest exposure potential to Cry protein in transgenic cotton fields are beneficial insects that feed on cotton pollen and nectar, particularly lepidopteran insects, and soil invertebrates. While EPA's risk assessments of Bt cotton have focused primarily on these taxa, BPPD recognizes that exposure to other nontarget organisms can occur and has required testing on representative species.

The EPA risk assessment is centered on adverse effects at the field exposure rates, which are typically based on protein expression levels within the plant for PIPs. Although it is recommended that non-target testing be conducted at a test dose 10X the EEC whenever possible, the test dose margin can be less than 10X where uncertainty in the system is low or where high concentrations of test material are not possible to achieve due to test organism feeding habits. BPPD may also allow for testing at lower doses in cases where many species are tested or tests are very sensitive, although the concentration used must exceed the EEC. For the purposes of the non-target organism studies submitted in support of Event T304-40 and Event GHB119, the test material dose levels were based on an estimated concentration of Cry1Ab or Cry2Ae protein expressed in the tissue(s) with the highest concentration based on protein expression studies (MRIDs 47634803, 47641903).       
      B. Ecological Effects Data for Event T304-40 and Event GHB119
	
In the absence of PIP-specific risk assessment guidance, EPA requires applicants for PIP registrations to meet the 40 CFR Part 158 data requirements for microbial toxins. These requirements include testing on birds, mammals, nontarget insects, honey bee, plants, and aquatic species, and information has been submitted to address these requirements. Limit dose testing on representative organisms from several taxa was performed in support of the T304-40 and GHB119 Section 3 FIFRA registrations. As stated above, BPPD's risk assessments focus heavily on beneficial nontarget invertebrates, since they are most closely related to organisms susceptible to the insecticidal action of Bt toxins. The Cry1Ab and Cry2Ae proteins are meant to target species within the order Lepidoptera (moths and butterflies). Bt toxins are known typically to have a limited host range. However, to address any unforeseen change in activity spectrum as a result of laboratory protein synthesis and to fulfill the published registration data requirements, EPA requires that test species used for non-target insect evaluations should include several invertebrate species that are not related to the target pests. Earthworm studies are also recommended. The toxicity of the Cry1Ab and Cry2Ae proteins has been evaluated on several species of invertebrates including the lady beetle, green lacewing, collembola, Daphnia, honey bee, and earthworm. Reproductive and/or developmental observations were also examined in the lady beetle, Daphnia, collembola, and honeybee studies. 

Since exposure may also occur to other nontarget organisms, EPA has received data to comply with the Agency's published non-target data requirements on other nontarget organisms. Tests to determine soil degradation of Cry1Ab and Cry2Ae proteins have also been submitted. The individual results for nontarget organism and soil degradation testing for Cry1Ab and Cry2Ae proteins are summarized in Tables 1 and 2, respectively. The studies are described in more detail below, and full reviews of each study for each event can be found in the individual Data Evaluation Records. 

 The October 2000 SAP recommended that while actual plant material is the preferred test material, bacteria-derived protein is also a valid test substance, particularly in scenarios where test animals do not normally consume cotton plant tissue and where large amounts of Cry protein are needed for maximum hazard dose testing. In support of the T304-40 and GHB119 registrations, test substances used in the submitted studies included bacteria-produced purified Cry1Ab and Cry2Ae protein, respectively. Comparative analyses have been performed to verify the equivalence of the bacterially-produced and purified Cry1Ab and Cry2Ae and the PIP Cry1Ab and Cry2Ae proteins produced in T304-40 or GHB119, respectively, and TwinLink(TM) cotton plants. Product characterization data (see Appendices A and B) demonstrate functional equivalence between the bacteria-derived Cry1Ab and Cry2Ae proteins used in toxicity testing and plant-expressed Cry1Ab and Cry2Ae proteins. The ecological risk assessment for the combined events expressed in TwinLink(TM) cotton is also dependent on equivalent expression levels of Cry1Ab and Cry2Ae protein in TwinLink(TM) cotton plants compared to the T304-40 and GHB119 parent lines. This bioequivalence has been demonstrated [see Product Characterization - Appendices A (Cry1Ab) and B (Cry2Ae and TwinLink(TM) Cotton)].
 
Table C.1. Summary of environmental effects studies and waiver justifications for Cry1Ab as expressed in T304-40 cotton submitted to comply with data requirements published in 40 CFR § 158.2150(d).

                               Data Requirement 
                                     OPPTS
                                   Guideline
                                Test Substance
   C.1. Cry1Ab PIP  -  Ecological Effects - Results Summary and Classification
                                   MRID No. 
Avian dietary testing, 
broiler chicken, Gallus domesticus 

885.4050
TwinLink(TM) (T304-40 x GHB119) cotton seed meal   
A 42-day dietary study showed no adverse affects to broiler chickens when fed a diet composed of 10% TwinLink(TM) cottonseed meal containing both Cry1Ab x Cry2Ae proteins. However, the presence of the proteins in the test material was not confirmed.
Classification:  Unacceptable
 47640011
Avian inhalation testing
885.4100

N/A
Acceptable waiver rationale
47871403
Avian oral testing, Northern bobwhite
Colinus virginianus
850.2100
N/A
Data waiver rationale was submitted to justify that testing is not necessary on the basis of lack of exposure and information on the toxicity of registered Cry proteins to birds.
Classification:  Acceptable
47903201
Wild mammal testing
885.4150

N/A
Acceptable bridging rationale to acute oral  toxicity test on laboratory mice
47871403
Freshwater fish testing, 
885.4200

N/A
Acceptable waiver rationale
47871403
Freshwater aquatic invertebrate testing, 
water flea, Daphnia  magna 
885.4240
 
Microbially-produced Cry1Ab
No adverse effects on survival or reproduction of Daphnia magna exposed to 48 ug Cry1Ab protein/L in a 10-day static renewal test. The 48-hr and 10-day EC50 values were determined to be > 48 ug Cry1Ab protein/L; the NOAEC is >= 48 ug Cry1Ab protein/L.
Classification:  Acceptable 
 47634819
Estuarine and marine animal testing 
885.4280

N/A
Acceptable waiver rationale
 47871403
Non-target plant testing
885.4300

N/A
Acceptable waiver rationale
 47871403
Non-target insect testing, pink-spotted lady beetle, Coleomegilla maculata 
885.4340

Microbially-produced Cry1Ab protein
No adverse effects on survival, time to pupation, and mean adult weight for Coleomegilla maculata larvae exposed to 19 ug Cry1Ab protein/g diet over 21 days. The LC50 was >19 ug Cry1Ab protein/g diet and the NOAEC is >= 19 ug Cry1Ab protein/g diet. 
Classification:  Acceptable
 47634815
Non-target insect testing, green lacewing,  Chyrsoperla rufilabris
885.4340
Microbially-produced Cry1Ab protein
No adverse effects on survival or pupation were observed in Chrysoperla rufilabris exposed to 20 ug Cry1Ab protein/g diet (nominal) in a 10-day laboratory bioassay. The nominal LC50 is >20 ug Cry1Ab protein/g diet and the NOAEC is >=20 ug Cry1Ab protein/g diet. 
Classification:  Acceptable
 47692701
Non-target insect testing, collembola, Folsomia candida
885.4340

Microbially-produced Cry1Ab protein
No adverse effects on survival or reproduction of collembola (Folsomia candida) exposed to 20 ug of Cry1Ab protein /g diet in a 28-day laboratory bioassay. The nominal LC50 was determined to be > 20 ug Cry1Ab protein/g diet and the nominal NOAEC is >=20 ug. 
Classification:  Acceptable
 47651106
Honeybee testing, Honeybee larvae,
Apis mellifera
885.4380

Microbially-produced Cry1Ab protein
No adverse effects on survival were observed in honey bee larvae treated in natal cells with 20 ug/L solution of Cry1Ab protein. The LC50 was determined to be > 20 ug Cry1Ab protein/mL.
Classification:  Acceptable
 47651105
Earthworm toxicity, 
Eisenia foetida
850.6200
N/A
Acceptable Waiver Rationale
 47634818
Soil fate and degradation
885.5200
Microbially-produced Cry1Ab protein
Based on the measured bioactivity in tobacco budworm (Heliothis virescens), the average DT50 for Cry1Ab protein in three different soils was 3.6 days.  
Classification:  Acceptable
 47634820


Table C.2. Summary of environmental effects studies and waiver justifications for GHB119 submitted to comply with data requirements published in 40 CFR § 158.2150 (d).
                               Data Requirement 
                                     OPPTS
                                   Guideline
                                Test Substance
                   C.2. Cry2Ae PIP - Ecological Effects  -  
                      Results Summary and Classification
                                   MRID No. 
Avian dietary testing, 
broiler chicken, Gallus domesticus 

885.4050
TwinLink(TM) (T304-40 x GHB119) cotton seed meal   
A 42-day dietary study showed no adverse affects to broiler chickens when fed a diet composed of 10% TwinLink(TM) cottonseed meal containing both Cry1Ab and Cry2Ae proteins. However, the presence of the proteins in the test material was not confirmed.
Classification:  Unacceptable
 47640011
 (same as above)
Avian inhalation testing
885.4100

N/A
Acceptable waiver rationale
47871503
Avian oral testing, Northern bobwhite
Colinus virginianus
850.2100
N/A



GHB119 cotton seed meal (raw)
Data waiver rationale was submitted to justify that testing is not necessary on the basis of lack of exposure.
Classification:  Supplemental
A 5-day feeding study with Northern bobwhite showed no adverse effects at 10% raw GHB119 cotton seed meal in the diet. 
Classification: Acceptable
47903301



48177501

Wild mammal testing
885.4150

N/A
Acceptable bridging rationale to acute oral  toxicity test on laboratory mice
47871503
Freshwater fish testing
885.4200

N/A 
Acceptable waiver rationale
47871503
Freshwater aquatic invertebrate, water flea neonate, Daphnia magna 
885.4240
 
Microbially-produced Cry2Ae protein
Daphnia magna were exposed to a nominal concentration of 330 ug Cry2Ae protein/L both in water and in food for 10 days. No adverse effects on immobilization/survival or reproduction were observed. The nominal 48-hr and 10-day EC50 values were >330 ug a.i./L and the nominal NOAEC was >= 330 ug a.i./L.
Classification:   Acceptable
 47641919
Estuarine and marine animal  testing
885.4280

N/A
Acceptable waiver rationale
 47871503
Non-target plant testing 
885.4300

N/A
Acceptable waiver rationale
 47871503
Non-target insect testing, pink-spotted lady beetle, Coleomegilla maculata 
885.4340

Microbially-produced Cry2Ae protein
No statistically significant effects on mortality, body weight, mean number of days to pupation, and mean number of days to adulthood were observed in Coleomegilla maculata exposed to 64 ug Cry2Ae protein/g of diet for 21 days. The nominal LC50 is > 64 ug Cry2Ae protein/g diet and the NOAEC is >= 64 ug Cry2Ae protein/g diet.
Classification:  Acceptable
 47641915
Non-target insect testing, green lacewing, Chrysoperla rufilabris
885.4340

Microbially-produced Cry2Ae protein
No statistically significant differences in mortality or % pupation were observed in green lacewing exposed to a nominal concentration of 10 or 110 ug Cry2Ae protein/g of diet for 10 days. The nominal LC50 is >110 ug Cry2Ae protein/g diet and the nominal NOAEC is >= 110 ug Cry2Ae protein/g diet.
Classification:  Acceptable
 47692501
Non-target insect testing, collembolan, Folsomia candida
885.4340

Microbially-produced Cry2Ae protein
Limit test:  The number of offspring per adult exposed to Cry2Ae protein was significantly reduced in Collembola exposed to 66 ug Cry2Ae protein/g diet for 28 days.  
Classification:  Supplemental
Definitive test:  No statistically significant differences in survival or reproduction among the test material and negative control groups were observed in Collembola exposed to 9.9, 20, 40, 80, or 160 ug of Cry2Ae protein/g diet. The nominal LC50 is >160 ug Cry2Ae protein/g diet and the nominal NOAEC is >=160 ug Cry2Ae protein/g diet.
Classification:  Acceptable
 47641917a
 47641917b
Earthworm toxicity, 
Eisenia foetida
850.6200
Microbially-produced
Cry2Ae protein
No adverse effects were observed on burrowing time, mortality, and weight change for earthworms (Eisenia fetida) were exposed to a nominal concentration of 100 mg Cry2Ae protein/kg of artificial soil for 14 days. The nominal 7- and 14-day LC50 was determined to be >100 mg a.i./kg, and the NOAEC was >=100 mg a.i./kg.
Classification:  Acceptable
 47641918
Honeybee  testing,  
Apis mellifera 
885.4380

Microbially-produced Cry2Ae protein
Larval survival did not differ significantly between the Cry2Ae protein treated group and the controls at capping and at emergence for honey bee (Apis mellifera) larvae exposed to 50 ug Cry2Ae protein/mL. Therefore, the LC50 is >50 ug Cry2Ae protein/mL. 
Classification:  Acceptable
 47641914
Soil fate and degradation
885.5200
Microbially-produced Cry2Ae protein
Based on the measured bioactivity in tobacco budworm (Heliothis virescens), the average DT50 for Cry2Ae protein in three different soils was 3.4 days. 
Classification:  Acceptable
 47641920

	





	1. Non-target Organism Study Summaries for T304-40 expressing Cry1Ab Protein

   	a.  Avian Wildlife
	
	Avian Dietary Toxicity

In a 42-day study (MRID 47640011), newly-hatched Ross #308 broiler chickens were fed formulated diet containing 10% TwinLink(TM) cottonseed meal  Control groups were fed formulated diet containing either 10% near isogenic cottonseed meal or 10% non-transgenic commercial cottonseed meal. There were no statistically significant treatment-related effects on survival, body weight, body weight gain, feed consumption, feed conversion ratio, or weight of chilled carcass, leg, thigh, wing, or breast among any of the test groups. DNA analysis of the prepared diets confirmed the presence of transgenic genetic material in the TwinLink(TM) diet, however the presence of the Cry1Ab and Cry2Ae proteins were not confirmed. The study report stated that the cottonseed meal was toasted, and this heating may have degraded the proteins. Thus, there is no assurance that the test animals were exposed to Cry1Ab or Cry2Ae proteins in the diet throughout the study, and the study cannot be used for avian risk assessment. A request to bridge to data from this study to the assessment of Cry1Ab protein expressed in T304-40 cotton (MRID 47871404) was not reviewed because the study was determined to be unacceptable.

	Avian Oral Toxicity  

Rationale was presented to justify waiving the avian oral toxicity data requirement for the Cry1Ab protein expressed in Event T304-40 cotton (MRID 47903201). The justification rested primarily on the expected minimal exposure of birds to Cry1Ab protein in T304-40 cotton as a result of their avoidance of cotton seed consumption, as well as the lack of observed effects of other Bt proteins that have been observed on birds (Mendelsohn et al. 2003). 

		Avian Inhalation Toxicity

Avian inhalation toxicity testing is typically not required for Bt cotton event registrations, since exposure to Cry proteins via inhalation is not expected. The applicant submitted data waiver rationale (MRID 47871403) that is consistent with BPPD's position on inhalation exposure for birds in Bt cotton, and this rationale is acceptable. An avian inhalation test is not required.

    	b. Wild mammals

BCS submitted acceptable data waiver rationale (MRID 47871403) to justify that testing with wild mammals is not necessary. An acute oral toxicity study with mice (MRID 47651103) has been submitted to EPA in support of the T304-40 registration, and in this study no mortalities, clinical signs of treatment-related effects, or treatment-related effects on body weight were observed on mice dosed at 2000 mg/kg body weight. This study has been determined to be acceptable. There is no indication that laboratory species would not be representative of wild mammals in this case; therefore, this information is suitable for use in assessment of risk to wild mammals, and no additional testing with wild species is necessary.      




   	c. Aquatic species

	Freshwater animals

Data waiver rationale (MRID 47871403) was submitted to fulfill the data requirements for freshwater fish for Cry1Ab protein in T304-40 cotton. Justification rested primarily on the lack of expected exposure of aquatic environments to Cry proteins produced in cotton, and this rationale was determined to be sufficient. Cotton pollen is widely known to be "sticky" (i.e., to form clumps) and too heavy to be dispersed by wind. As a result, little pollen is expected to move off field and enter aquatic systems. BPPD has also concluded that little Bt cotton plant matter is expected to move off field (e.g., see U.S. EPA 2008a). Therefore, exposure in aquatic systems is expected to be low, and testing of freshwater fish is determined to be unnecessary.    

Because of concerns raised in a study with the leaf shredding (caddis fly) trichopteran, Lepidostoma liba (Rosi-Marshall, et al. 2007), EPA is  requiring applicants of Bt corn and cotton to submit 7-10 day freshwater invertebrate toxicity studies to reduce uncertainty regarding this issue. A 10-day study with Daphnia is available with which to determine the risk to freshwater invertebrates as a result of exposure to Cry1Ab protein in T304-40 cotton. Daphnia were exposed to a nominal concentration of 48 ug Cry1Ab protein/L, which is based on the mean expression level of protein present in one kilogram of pollen, dispersed in one liter of water (see MRID 47634803). At this concentration, with exposure in both water and in food treated with Cry1Ab protein, no mortalities and no significant differences in the number of offspring produced were observed. Therefore, the EC50 was determined to be > 48 ug Cry1Ab protein/L, and the NOAEC was determined to be >= 48 ug Cry1Ab protein/L.    

	Estuarine and marine animals

BPPD typically waives toxicity studies with estuarine/marine fish and invertebrates based on an expected minimum exposure in these environments; however, the applicant submitted data waiver rationale (MRID 47871403) to fulfill data requirements for estuarine and marine animals. The waiver rationale presented was the same as that discussed above for freshwater fish and was determined to be sufficient based on BPPD's previous conclusions about exposure to Cry proteins in aquatic environments.  

   	d. Terrestrial and aquatic plant species

BPPD typically waives nontarget plant testing for Bt Cry proteins, since the active ingredient is an insect toxin (Bt -endotoxin) that has never shown any toxicity to plants. The applicant submitted acceptable data waiver rationale (MRID 47871403) that was consistent with BPPD's position on nontarget plant risks to Bt -endotoxin.  
	
   	e. Invertebrate species        

In a study with the ladybird beetle (Coleomegilla maculata, MRID 47634815), larvae were exposed to a nominal concentration of 19 ug Cry1Ab protein/g of artificial diet for 21 days. This amount is meant to represent 10X the mean expression level of Cry1Ab protein in the cotton plants. Larval and adult mortality, time to pupation, time to adulthood, and adult weight did not differ significantly from the negative control group. Larval mortality was 7.5% and 12.5% in the treated and control groups, respectively; adult mortality was 0% and 2.9% in the treated and control groups, respectively. The nominal LC50 was determined to be >19 ug Cry1Ab protein/g of diet and the NOAEC was determined to be >= 19 ug Cry1Ab protein/g of diet.

A study with green lacewing (Chrysoperla rufilabris, MRID 47692701) larvae exposed 25 individuals to a nominal concentration of 20 ug Cry1Ab protein/g of artificial diet for 10 days. The amount to which the larvae were exposed throughout the test was meant to represent 10X the mean fresh weight expression level of Cry1Ab protein in T304-40 cotton plants. Mortality and pupation measured in the test group (94% and 76%, respectively) did not differ significantly from the negative control (96% and 76%, respectively). Based on the results, the LC50 for Cry1Ab protein in C. rufilabris is >20 ug a.i./g and the NOAEC was determined to be >= 20 ug Cry1Ab protein /g of diet.
 
Collembola (Folsomia candida) are soil dwelling insects that consume detritus. In a 28-day study (MRID 47651106), 40 collembola were exposed to 20 ug Cry1Ab protein/g diet and observed for mortality and number of offspring produced per adult. Mortality of the collembola was 25% for the groups exposed to Cry1Ab protein and 12.5% for the control groups. The number of offspring produced per adult was 46.9 and 43.9 for the Cry1Ab treated and control groups, respectively. Neither of these endpoints differed significantly between treated and control groups. Based on the results, the, the LC50 based on the nominal concentration is > 20 ug Cry1Ab protein/g diet and the NOAEC was established to be >= 20 ug Cry1Ab protein/g diet. 

Data waiver rationale (MRID 47634818) was presented to justify that toxicity testing for Cry1Ab protein as expressed in T304-40 cotton is not necessary with the earthworm. The 14-day LC50 for earthworms exposed to Cry1Ab protein derived from E. coli was previously determined to be > 200 mg/kg of artificial soil (U.S. EPA 2010, Mendelsohn et al. 2003). Since there is no evidence of toxicity of Cry1Ab protein to earthworms, BPPD determined that additional testing with earthworms is not needed.

A study with honey bee (Apis mellifera) was also performed. In this study, 2-3 day old honey bee larvae were exposed to a 10 uL droplet of solution containing 20 ug/mL Cry1Ab protein, which was intended to represent >10X the mean fresh weight expression concentration of Cry1Ab protein in T304-40 cotton leaves, and > 400 times the exposure to the amount of Cry1Ab protein expressed in T304-40 pollen. Mortality at capping and at emergence was observed, and neither of these differed significantly from controls. Emergence was considered to be complete on Day 16. Of the larvae treated with Cry1Ab protein, 85.25% survived to capping, and all of these survived to emergence; 82.5% of the controls survived to capping and all of these also survived to emergence. Therefore, the LC50 was determined to be > 20 ug Cry1Ab protein/mL.

      2. Non-target Organism Study Summaries for GHB119 expressing Cry2Ae protein

   	a.  Avian Wildlife

	Avian Dietary Toxicity

The broiler chicken dietary study described in the summaries above for T304-40 (MRID 47640011) was also submitted for this data requirement for Cry2Ae protein expressed in GHB119 cotton. Since there is no assurance that the test animals were exposed to Cry1Ab or Cry2Ae protein in the diet throughout the study, it cannot be used for avian risk assessment for Cry2Ae protein. A request to bridge to data from this study to the assessment of Cry2Ae protein expressed in GHB119 cotton (MRID 47871504) was not reviewed because the study was determined to be unacceptable.
	Avian Oral Toxicity  

Rationale was presented to justify waiving the avian oral toxicity data requirement for the Cry2Ae protein expressed in Event GHB119 cotton (MRID 47903301). The justification rested primarily on the expected minimal exposure of birds to Cry2Ae protein in GHB119 cotton as a result of their avoidance of cotton seed consumption, as well as the lack of observed effects of other Bt proteins that have been observed on birds. However, this rationale does not account for other routes of oral exposure in birds (e.g., consumption of insects). 

BCS submitted a feeding study in which Northern bobwhite were exposed to raw Event GHB119 cotton seed meal (containing Cry2Ae protein) at 10% of their daily diet for five days (MRID 48177501). No mortality or behavioral abnormalities were observed, and significant differences in body weight gain or food consumption were not detected. This study is sufficient to show that hazard to birds exposed to the dietary concentration tested (up to 0.241 ppm diet or 10% of GHB119 cottonseed) is not expected. This study does not sufficiently test hazard at higher dietary concentrations. EPA has previously determined that an exposure level of 10% raw cottonseed meal containing Cry protein is a sufficient exposure level for determining the effects of Cry proteins expressed in cotton in birds (e.g., U.S. EPA 2005a, U.S. EPA 2005b, U.S. EPA 2008). Based on these exposure assumptions, the study was determined to be acceptable; however, there is uncertainty about the potential effects to birds if field exposure occurs at higher concentrations. It is noted that no avian toxicity data submitted to the Agency to date has shown detrimental effects of Cry proteins on birds, including data on the Cry2Ab2 protein (U.S. EPA 2005a). Therefore, BPPD concludes that exposure to Cry2Ae protein is not likely to cause adverse effects in birds. However, a study in which birds are exposed to higher doses of Cry2Ae protein would reduce uncertainties related to actual field exposure.  

		Avian Inhalation Toxicity

Avian inhalation toxicity testing is typically not required for Bt cotton event registrations, since exposure to Cry proteins via inhalation is not expected. The applicant submitted data waiver rationale (MRID 47871503) that is consistent with BPPD's position on inhalation exposure for birds in Bt cotton, and this rationale is acceptable. An avian inhalation test is not required.

    	b. Wild mammals

The applicant submitted acceptable data waiver rationale (MRID 47871503) to justify that testing with wild mammals is not necessary. An acute oral toxicity study with mice (MRID 47076902) has been submitted to EPA in support of the GHB119 registration, and in this study no mortalities or treatment-related signs of adverse effects were observed on mice dosed at 2000 mg/kg body weight. This study was determined to be acceptable. This information is suitable for use in assessment of risk to wild mammals, and no additional testing with wild species is necessary at this time.

		c. Aquatic animals

Data waiver rationale (MRID 47871503) was submitted to fulfill the data requirements for freshwater fish for Cry2Ae protein in GHB119 cotton. Justification rested primarily on the lack of expected exposure of aquatic environments to Cry proteins produced in cotton, and this rationale was determined to be sufficient. The same rationale as presented above for Cry1Ab protein was presented for Cry2Ae protein, and this is consistent with BPPD's previous conclusions that little aquatic exposure is expected from the use of cotton expressing Cry proteins. Since BPPD has concluded that exposure in aquatic systems is expected to be very low, testing of freshwater fish is unnecessary.    

A 10-day study with Daphnia magna is available with which to determine the risk to freshwater invertebrates as a result of exposure to Cry1Ab protein in T304-40 cotton. Daphnia were exposed to a nominal concentration of 330 ug Cry2Ae protein/L contained in water alone and water and food. This exposure level is based on the fresh weight expression level of protein present in one kilogram of pollen (see MRID 47641933), dispersed in one liter of water. No mortalities and no significant differences in the number of offspring produced were observed. Therefore, the EC50 was determined to be > 330 ug Cry2Ae protein/L, and the NOAEC was determined to be >= 330 ug Cry2Ae protein/L.    

		 Estuarine and marine animals

BPPD typically waives toxicity studies with estuarine/marine fish and invertebrates; however, the applicant submitted data waiver rationale (MRID 47871503) to fulfill data requirements for estuarine and marine animals. The waiver rationale presented was the same as that discussed above for exposure of freshwater fish to Cry1Ab protein and was determined to be sufficient based on BPPD's conclusions about exposure in aquatic environments.  

		d. Terrestrial and aquatic plant species

BPPD typically waives nontarget plant testing for Bt Cry proteins, since the active ingredient is an insect toxin (Bt -endotoxin) that has never shown any toxicity to plants. The applicant submitted acceptable data waiver rationale (MRID 47871503) that was consistent with BPPD's position on nontarget plant risks to Bt -endotoxin. 
		
   	e. Invertebrate species        

A study with ladybird beetle (Coleomegilla maculata, MRID 47641915) larvae exposed 40 individuals to a nominal concentration of 64 ug Cry1Ab protein/g of artificial diet for 21 days. This amount was approximately 7X the fresh weight expression level of Cry2Ae protein in the leaves of GHB119 cotton plants (see MRID 47641903). Larval and adult mortality, time to pupation, time to adulthood, and adult body weight did not differ significantly from the negative control group. Larval mortality was 5.0% and 12.8% in the treated and corresponding control group, respectively; adult mortality was 0% and 3.3% in the treated and control groups, respectively. The nominal LC50 was determined to be >64 ug Cry1Ab protein/g diet and the NOAEC based on development and body weight is >= 64 ug Cry2Ae protein/g diet.

A study with green lacewing (Chrysoperla rufilabris, MRID 47692501) larvae exposed 25 individuals to a nominal concentration of 10 ug and 100 ug Cry2Ae protein/g of artificial diet for 10 days. These concentrations were meant to test effects at 1X and 10X the fresh weight expression level of Cry2Ae protein as measured in the leaves of GHB119 cotton plants (see MRID 47641903). Mortality and percent pupation were measured and compared to negative and buffer controls, and no significant differences were observed. Survival of the 10 ug and 100 ug Cry2Ae protein/g diet test groups and the buffer controls was 100%, 96%, and 100%, respectively. Pupation in these groups was 84%, 72%, and 88%, respectively. The results establish a nominal LC50 of >100 ug Cry2Ae protein/g diet and a nominal NOAEC of >= 100 ug Cry2Ae protein/g diet. 

Collembola (Folsomia candida) are soil dwelling insects that consume detritus. In a 28-day single-dose study (MRID 47641917a), collembola were exposed to 66 ug Cry2Ae protein/g diet and observed for mortality and number of offspring produced per adult. Mortality of the collembola was 27.5% for the groups exposed to Cry2Ae protein and 22.5% for the control groups. The number of offspring produced per adult was 10.3 and 44.5 for the Cry2Ae protein treated and control groups, respectively, and the difference was statistically significant. Because of this outcome, a definitive test was performed on collembolan exposed to nominal concentrations of 9.9, 20, 40, 80, or 160 ug Cry2Ae protein/g diet (MRID 47641917b). Negative and positive controls were also included, and a bioassay on a sensitive species (Heliothis virescens) confirmed bioactivity. Mortality of the test groups from lowest to highest concentration was 2.5%, 0%, 0%, 13%, and 0%, and mortality in the negative controls was 7.5%. Mean number of offspring did not differ significantly among any of these groups. With this test, the nominal LC50 was determined to be >160 ug Cry2Ae protein/g diet and the NOAEC was determined to be >=160 ug Cry2Ae protein/g diet.  

A 14-day earthworm toxicity study (MRID 47641918) was submitted, wherein earthworms were exposed to 100 mg Cry2Ae protein/gram of artificial soil. This concentration was intended to be approximately 10X the fresh weight concentration of Cry2Ae protein expressed in the leaves of GHB119 cotton plants (see MRID 47641903). The average burrowing time was equal for both the treated and control groups, and no mortalities were observed in either group. Mean weight decreased by 22% in the control group and by 20% in the test material group, which was expected since the artificial soil did not contain nutritive material. Based on the results the nominal LC50 was determined to be >100 mg Cry2Ae protein/kg, and the NOAEC was >=100 mg Cry2Ae protein/kg.
 
A study with honey bee (Apis mellifera) was also performed. In this study, 2-3 day old honey bee larvae were exposed to a 10 uL droplet of solution containing 50 ug/mL Cry1Ab protein, which was intended to represent >50X the exposure to the amount of Cry2Ae protein expressed in GHB119 cotton pollen (assuming the test solution is approximately the same weight, see MRID 47641913). Mortality at capping and at emergence were observed, and neither of these differed significantly from controls. Emergence was considered to be complete on Day 16. Of the larvae treated with Cry2Ae protein, 93.75% survived to capping, and all of these survived to emergence; 95% of the controls survived to capping and all of these also survived to emergence. Therefore, the LC50 is >50 ug Cry2Ae protein/mL.

      3. Soil Fate Studies for Cry1Ab and Cry2Ae PIPs

Soil organisms may be exposed to Cry1Ab and Cry2Ae protein through contact with cotton plant roots (by direct feeding), cotton plant root exudates, incorporation of above-ground plant tissues into soil following harvest, or by soil-deposited pollen. Some evidence suggests that soils which are high in clays and humic acids are more likely to bind Cry protein. However, neutral pH soils tend to have high microbial activity and microbes contribute to Cry protein degradation. In addition, a study on the release of Cry proteins in the root exudates of Bt cotton has shown that no Cry proteins were detected immunologically or by larvicidal assay in any soil or hydroponic solution in which Bt cotton had been grown (Saxena and Stotzky, 2001). The weight of evidence indicates that Cry proteins do not accumulate in soil to arthropod-toxic levels. Because Cry proteins are toxins derived from soil-inhabiting bacteria, Bacillus thuringiensis, and found in commercial microbial insecticides (De Maagd et al., 2003 and Graser and Song, 2006), the degradation of Cry1Ab and Cry2Ae from Event T304-40 and Event GHB119 cotton is expected to be similar to that of these proteins that are produced naturally from soil bacteria. Nonetheless, BPPD required the following soil fate evaluations to support the Event T304-40 and GHB119 Bt cotton registrations. These studies are summarized here, and detailed evaluations are provided in their associated DERs.

A laboratory aerobic soil degradation study was conducted to determine the DT50 (time for 50% of the initial concentration of bioactive material to dissipate) for Cry1Ab protein. Soil used in the study was collected from three locations throughout the region where transgenic cotton is grown, and each was of a different texture (Proctor, AR: clay loam; Senatobia, MS: silt loam; East Bernard, TX: sandy clay loam). Each soil was treated once with 15 ug/g bacterially-produced Cry1Ab protein, and then sampled periodically over 45 days. These samples were used in sensitive insect bioassays. The bioassay used tobacco budworm (Heliothis virescens) larvae exposed to the soil samples incorporated into lepidopteran diet. A concurrent six-day dose response study with H. virescens larvae exposed to 0.004 to 0.28 ug/g Cry1Ab protein in the soils was conducted to generate a dose-response curve to determine the degradation of Cry1Ab protein over time based on the bioactivity response. The DT50 was determined from the Cry1Ab protein concentration first-order exponential regression curve. The average DT50 for the three soils was 3.6 days. The soils were also analyzed for Cry1Ab protein content by enzyme linked immunosorbent assay (ELISA), which showed that Cry1Ab protein steadily degraded in soil during the study.  

A similar study was conducted to determine the DT50 for Cry2Ae protein in soil. Soils were collected from the same locations as above for the study with Cry1Ab protein. The soils were treated once with 50 ug/g Cry2Ae protein and then sampled periodically over 45 days. A sensitive insect bioassay using the same species (Heliothis virescens) was conducted using identical methods as with Cry1Ab protein, except that the concurrent six-day dose response study exposed H. virescens larvae to 0.056 to 3.4 ug/g Cry2Ae protein to generate a standard curve for determination of Cry2Ae protein concentration based on bioactivity response. Based on the bioassays, the average DT50 was determined to be 3.4 days. The soils were also analyzed for Cry2Ae protein content by ELISA, which showed that Cry2Ae protein steadily degraded in soil during the study.

These studies utilized field soil spiked with purified insecticidal protein produced by bacteria. This approach is useful because dose responses can be easily quantified. However, the degradation and accumulation of Bt Cry proteins found within decaying plant tissue may behave differently than proteins in artificially spiked soil. Thus, the presence of low levels of Bt Cry proteins in the soil (at or below the level of detection) is anticipated until all plant tissue is degraded. However, the reviewed data show that Cry1Ab and Cry2Ae proteins will be quickly degraded upon release from decaying plant tissue. 

Based on FIFRA Scientific Advisory Panel recommendations and public comments, the Agency has required three year soil fate studies for the currently registered Cry protein producing crops grown in a variety of soils and environmental conditions, as a condition of registration. More recently, a comprehensive review of all available scientific data on ecological effects of commercially grown GM crops over ten years was completed (Sanvido, et al. 2007). The review concluded "none of the laboratory or field studies suggest accumulation of Bt-toxins in soil over several years of cultivation" and "experience from commercial cultivation indicates that Bt-toxin will not persist for long periods under natural conditions."  The Agency agrees with these conclusions. Collectively, the long-term field studies for Bt crops also confirm the previous SAP conclusion that "bioaccumulation is not expected to occur with transgenic proteins because biodegradation mechanisms for proteins are ubiquitous" (US EPA, 2000). More importantly, the numerous laboratory studies that demonstrated rapid protein degradation in soil of Bt proteins produced in Bt crops (when performed under realistic environmental conditions) can be considered predictive that Bt protein in soil is not likely to persist or accumulate in soil after continuous cultivation. 

In light of these published findings and the rapid degradation of Cry1Ab and Cry2Ae proteins in soil as demonstrated in the insect bioassays described above, there is no indication that the proteins expressed in Event T304-40 and Event GHB119 cotton are likely to persist in soil after continuous cultivation. Therefore, no additional long-term field studies are required for these PIP products.

	4. Supplemental Information

BCS submitted summaries of nontarget organism testing with Cry1Ab and Cry2Ae proteins (MRIDs 47634813 and 47641913, respectively), which summarized the information above, and included an assessment of ecological risk and some discussion of environmental fate. These summaries were determined to be supplemental, but provided little additional information for use in the risk assessment.   

      C. Environmental Risk Assessment for Event T304-40 and Event GHB119

		1. Effects to Nontarget Wildlife, Invertebrates, and Plants

   		a.  Avian Wildlife

Birds are known to forage in cotton fields. While birds are unlikely to consume cotton plant material, particularly cotton seeds, due to high concentrations of gossypol, they could be exposed to Cry1Ab and/or Cry2Ae proteins through consumption of invertebrates that feed on Bt cotton plant tissues. Exposure is expected to be low, and data and rationale were sufficient to conclude that Cry1Ab and Cry2Ae proteins are not expected to cause adverse effects in birds. There is some uncertainty related to the exposure level in the testing with Cry2Ae protein compared to field exposures, but EPA previously determined that the exposure level in the study was sufficient for avian testing. Testing at higher doses would reduce this uncertainty. 

    		b. Wild mammals

Wild mammals may be exposed to Cry1Ab protein in Event T304-40 cotton or Cry2Ae protein in Event GHB119 cotton by consuming plant parts and/or invertebrates that feed on the cotton plants. Toxicity testing with laboratory mice showed no adverse of Cry1Ab or Cry2Ae proteins at doses of 2000 mg/kg bodyweight each. This amount is much higher than concentrations that wild mammals are expected to encounter in the field. Therefore, based on this information, adverse effects to wild mammals are not expected as a result of the use of Event T304-40 and Event GHB119 cotton. 

		c. Freshwater animals

BPPD has concluded that exposure in the aquatic environment to the Cry proteins in Bt cotton is expected to be low. Cotton plants are not wind-pollinated, and cotton pollen is widely understood to be spiny and "sticky," leading to the formation of clumps of pollen that are not easily dispersed by wind. As a result, the majority of the cotton pollen that is produced is not expected to travel very far off-field. BPPD has previously determined that exposure in aquatic environments resulting from deposition of other cotton plant tissues is minimal (e.g., see U.S. EPA 2008a). Therefore, exposure of Cry1Ab or Cry2Ae protein from use of Event T304-40 or Event GHB119 cotton, respectively, to freshwater fish and invertebrates is expected to be minimal, and adverse effects to these species are not expected to occur.

		d. Estuarine and marine animals

Significant amounts of Cry1Ab and Cry2Ae protein resulting from Event T304-40 or GHB119 cotton use, respectively, are not expected to reach marine/estuarine areas. BPPD concludes that adverse effects to fish and invertebrates in these environments are not expected. 

   		e. Terrestrial and aquatic plant species

BPPD typically waives nontarget plant testing for Bt Cry proteins, since the active ingredient is an insect toxin (Bt -endotoxin) that has never shown any toxicity to plants. BPPD has concluded that adverse effects to terrestrial and aquatic plants are not anticipated.
	
   		f. Invertebrate species        

Nontarget insects are expected to receive exposure primarily through consumption of Bt cotton pollen and/or nectar, though they may also consume pest insects that feed on cotton plant tissue or rarely consume other cotton plant tissues themselves. The principal route of soil-dwelling invertebrates, such as collembola, earthworms, and rove beetles, is assumed to be from decomposing plant tissue and plant exudates in soil. Studies submitted to the Agency with Cry1Ab and Cry2Ae proteins in lady bird beetle, green lacewing, collembola, and earthworm showed no adverse effects at concentrations well above those that would be encountered by nontarget insects in the environment. Therefore, BPPD does not anticipate the occurrence of adverse effects to nontarget invertebrates as a result of the use of T304-40 or GHB119 cotton. Additionally, studies with honey bee larvae also showed no adverse effects to honey bees at exposure levels to Cry1Ab and Cry2Ae proteins above those expected to be encountered in the environment. Thus, risk to honey bees as a result of the proposed registrations is also expected to be minimal. 

It should be noted that effects to invertebrates that are recognized as cotton pests (i.e., insects that feed detrimentally on cotton plants) that are not targeted by the Cry1Ab and Cry2Ae proteins (but are controlled by other means) are not included in BPPD's assessment of risk to nontarget insects. 

		2. Effects on Soil Microorganisms 

Numerous published studies indicate that exposure to Cry protein produced in Bt PIP crop plants does not adversely affect soil microorganisms (Sanvido et al., 2007). Although a minimal transient increase and shift in microbial populations may result from the presence of transgenic plant tissue in soil, no adverse effects have been attributed to the Cry protein. In addition, the soil degradation studies above show that the Cry1Ab and Cry2Ae proteins degrade quickly in soil. 

In addition, there are several ongoing U.S. Department of Agriculture and EPA Office of Research and Development funded research projects evaluating the effects of Cry protein crops on soil microbial flora. If adverse effects are seen from this or any other research, the Agency will take appropriate action to mitigate potential risks (U.S. EPA 2010).

With regard to the impact of genetically engineered crops on soil, it is important to note that agricultural practices themselves cause large changes in soil and soil microbial composition. Furthermore, factors such variations in seasons and weather, plant growth stage, and plant varieties, independent of being genetically engineered, are also responsible for significant shifts in soil microbial communities. Most studies with genetically engineered crops to date have shown minor or no effects on soil microbes beyond the variation caused by the factors listed above. 

		3. Horizontal Transfer of Transgenes from Bt Crops to Soil Organisms 

The EPA has evaluated the potential for horizontal gene transfer (HGT) from Bt crops to soil organisms and has considered possible risk implications if such a transfer were to occur. Genes that have been engineered into Bt crops are mostly found, or have their origin, in soil-inhabiting bacteria. Soil is also the habitat of other toxin-producing bacteria, and transfer of these genes and/or toxins to other microorganisms or plants has not been detected. Furthermore, several published experiments, that were conducted to assess the likelihood of HGT, have been unable to detect gene transfer under typical environmental conditions. Horizontal gene transfer to soil organisms has only been detected with very promiscuous microbes under laboratory conditions designed to favor transfer. 

As a result of these findings and the fact that the Bt toxins engineered into T304-40 and GHB119 were derived from soil-inhabiting bacteria, the EPA has concluded that the risk of HGT of transgenes found in Cry1Ab or Cry2Ae producing cotton is low.

		4. Gene Flow and Weediness Potential 

Movement of transgenes from crop plants into weeds is a significant concern, due to uncertainty regarding the effect that a new pest resistance gene may have on plant populations in the wild. Under FIFRA, the Agency has reviewed the potential for gene capture and expression of Cry proteins in commercial Bt cotton by wild or weedy relatives of cotton in the United States, its possessions or territories. 

There is a possibility for gene transfer in locations where wild or feral cotton relatives exist. Therefore, EPA has required stringent sales and distribution restrictions on Bt cotton within these areas to prevent outcrossing or hybridization from the crop to sexually compatible relatives. There are only four areas in the United States and its territories wherein cultivated cotton has the opportunity to outcross to wild or feral species, which are genetically compatible: (1) southern Arizona, (2) Hawaiian islands, (3) southern Florida and 4) Puerto Rico. G. thurberi (Arizona Wild Cotton) is present in the elevated regions of Arizona and does not grow in areas of commercial cotton production. G. thurberi is a diploid and produces sterile, triploid progeny when crossed with the tetraploids G. hirsutum or G. barbadense. In the very south of Florida, feral G. hirsutum exists in apparently self-sustaining populations. Since these would readily cross with cultivated cotton, sale of Bt-Cotton is restricted south of Interstate 60. There is currently no commercial cotton production in the southern part of Florida. Evidence from germplasm collections indicates that feral G. barbadense and possibly G. hirsutum exist in the U.S. Virgin Islands. There is presently no production of commercial cotton in either of these places; hence, outcrossing is not currently an issue in these areas. For a detailed review of the Agency's assessment of the potential for gene capture and expression of Bt endotoxins by wild or weedy relatives of cotton in the U.S., its possessions or territories, see the EPA Biopesticides Registration Action Document (BRAD) for the Bacillus thuringiensis (Bt) Plant-Incorporated Protectants, dated October 15, 2001.

		5. Conclusions

BPPD concludes that significant adverse effects on birds, mammals, nontarget insects, honey bees, freshwater and marine/estuarine fish and invertebrates, and terrestrial and aquatic plants are not expected as a result of the use of Event T304-40 and Event GHB119 in cotton. EPA has also determined that there is no significant risk of gene capture and expression of Cry1Ab or Cry2Ae protein by wild or weedy relatives of cotton within the U.S. or its territories as long as the current sales and distribution restrictions on Bt cotton are in place. Available data do not indicate that Cry proteins have any measurable adverse effect on microbial populations in the soil, nor has horizontal transfer of genes from transgenic plants to soil bacteria been demonstrated.   
      
      6. Impacts on Endangered Species 

Because of the selectivity of Cry1Ab and Cry2Ae proteins for lepidopteran species, endangered species concerns are also expressed for insect species in the order Lepidoptera. The Agency has previously determined that any potential concern regarding range overlap of endangered Lepidoptera with cotton production was mainly restricted to the Kern primrose sphinx moth (Euproserpinus euterpe). However, the Agency determined that cotton is not a host plant for this species nor do host-range considerations for this species or other endangered insects place habitat in or near cotton fields (US EPA 2008a).     

As previously noted, there is a possibility for gene transfer in locations where wild or feral cotton relatives exist. As a result, EPA requires stringent sales and distribution restrictions on Bt cotton within these areas to preclude outcrossing or hybridization from the crop to sexually compatible relatives. Therefore, EPA does not expect that any threatened or endangered species will be affected by outcrossing to wild relatives or by competition with such entities.

Since Cry1Ab protein has not been shown to have toxic effects on mammals, birds, freshwater invertebrates, terrestrial nontarget invertebrates, and plants, and exposure is not anticipated in aquatic environments. Therefore, the Agency makes a "No Effect" determination for direct and indirect effects to listed species of these taxa for Cry1Ab protein expressed in Event T304-40 cotton. 

Since Cry2Ae protein has not been shown to have toxic effects on birds, mammals, freshwater invertebrates, terrestrial nontarget invertebrates, and plants, and also because exposure is not anticipated in aquatic environments, the Agency makes a "No Effect" determination for direct and indirect effects to listed species of these taxa for Cry2Ae protein expressed in Event GHB119 cotton. 

      D. Data Needed to Confirm T304-40 and GHB119 Non-Target Hazard Assessment

BPPD has sufficient information to believe that there is no significant risk from the proposed uses of Event T304-40 and Event GHB119 cotton to non-target avian and mammalian wildlife, terrestrial invertebrates, fish and aquatic invertebrates, and plants and soil organisms. As described above, uncertainty related to testing with Cry2Ae protein in birds would be reduced by additional testing at higher doses. 

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
II.  Environmental Risk Assessment for Cry1Ab and Cry2Ae proteins expressed in T304-40 X GHB119 (TwinLink(TM)) Cotton Hybrid 
--------------------------------------------------------------------------------

      A. Background 

BCS developed T304-40 x GHB119 (TwinLink(TM)) cotton by conventional breeding of transgenic Event T304-40 cotton and transgenic Event GHB119 cotton, which express the Cry1Ab and Cry2Ae insecticidal proteins, respectively, for control of certain lepidopteran pests. Event T304-40 expresses a cry1Ab gene, which is derived from the soil bacterium, Bacillus thuringiensis (Bt) subsp. berliner, and Event GHB119 expresses a cry2Ae gene derived from Bt subsp. dakota. 

	B. Ecological effects data for T304-40 x GHB119 (TwinLink(TM)) cotton hybrid

Environmental effects data are available for studies conducted with Cry1Ab and Cry2Ae proteins. These can be used in an assessment of risk for TwinLink(TM) cotton, provided there is no evidence of a synergistic association between Cry1Ab and Cry2Ae protein and protein expression levels of Cry1Ab and Cry2Ae are equivalent between the T304-40 and GHB119 parent lines and the TwinLink(TM) hybrid. Data have been submitted with which to determine the potential for synergy and protein expression levels. Additional toxicity data have been submitted in which the combination of these proteins has been tested (TwinLink(TM) plant material was used as the test substance) with some nontarget species, including honeybee, collembola, and broiler chicken. Additionally, data waiver rationales were presented for freshwater fish, marine/estuarine fish and invertebrates, nontarget plants, avian oral and inhalation toxicity, and wild mammals. Data available on TwinLink(TM) cotton are summarized in Table C.3, and descriptions of these studies and waiver rationales are provided below. Additional review of these studies is provided in their associated DERs.

For most studies required, BCS is seeking to bridge from data generated on the proteins expressed individually in the parent lines to the combined event. Data showing a lack of synergism between the Cry1Ab and Cry2Ae proteins are necessary for bridging, as well as data showing equivalent expression of the proteins within the TwinLink(TM) plants and the individual parent lines. A synergism study was submitted and is discussed below. Studies showing equivalence were reviewed and found acceptable in the product characterization review for TwinLink(TM) (BRAD Appendices A and B). BPPD concluded that the data from the expression of the proteins in the parent lines can be bridged to those expressed in the combined event.


Table C.3. Summary of environmental effects studies and waiver justifications for T304-40 x GHB119 (TwinLink(TM)) submitted to comply with data requirements published in 40 CFR § 158.2150 (d).
                               Data Requirement 
                                     OPPTS
                                   Guideline
                                Test Substance
  C.3. TwinLink(TM) - Ecological Effects - Results Summary and Classification 
                                   MRID No. 
Avian dietary testing, 
broiler chicken, Gallus domesticus 

885.4050
TwinLink(TM) (T304-40 x GHB119) cotton seed meal   
A 42-day dietary study showed no adverse affects to broiler chickens when fed a diet composed of 10% TwinLink(TM) cottonseed meal containing both Cry1Ab and Cry2Ae protein. However, the presence of the proteins in the test material was not confirmed.
Classification:  Unacceptable
 47640011
Avian inhalation testing
885.4100

N/A
Acceptable waiver rationale
47871603
Avian oral testing, Northern bobwhite
Colinus virginianus
850.2100
N/A
Data waiver rationale was submitted to justify that testing is not necessary on the basis of lack of exposure.
Classification:  Supplemental (subsequent testing with Cry2Ae protein was performed)
47903101
Freshwater fish testing, 
885.4200

N/A
Acceptable waiver rationale
47871603
Estuarine and marine animal testing 
885.4280

N/A
Acceptable waiver rationale
 47871603
Non-target plant testing
885.4300

N/A
Acceptable waiver rationale
 47871603
Non-target insect testing, collembola, Folsomia candida
885.4340

Lyophilized TwinLink(TM) cotton leaves
Collembola (Folsomia candida) were fed diet containing nominal concentrations of 0, 5, 20, or 50% lyophilized TwinLink(TM) cotton leaf, or 50% lyophilized near isoline cotton leaf for 28 days. No statistically significant differences in survival or reproduction among the test material and control groups were observed, indicating that exposure to TwinLink(TM) plant tissue at the levels tested has no effect on survival or reproduction of Collembola.
Classification:  Acceptable
 47640010
Honeybee testing, Honeybee larvae,
Apis mellifera
885.4380

TwinLink(TM) pollen
Honey bee larval cells were treated with 10 uL of a solution containing 2 mg of TwinLink(TM) pollen or near isoline pollen and observed for survival up to the time of emergence. A significant difference in survival between these two groups was not observed, indicating a lack of adverse effects of TwinLink(TM) pollen at the tested level on honey bee larvae. 
Classification:  Acceptable
 47640009
Synergism
N/A

Sensitive insect dose-response bioassays with T304-40, GHB119, and TwinLink(TM) plant material showed that Cry1Ab and Cry2Ae proteins expressed in TwinLink(TM) plants are not synergistic. 
Classification:  Acceptable
 48480009


	1. Nontarget Organism Study Summaries for TwinLink(TM)

   	a. Avian Wildlife

	Avian Dietary Toxicity

The broiler chicken feeding study (MRID 47640011) discussed above in the assessment for Events T304-40 and GHB119 was also submitted to provide data on toxicity for TwinLink(TM). As stated above, since there is no assurance that the test animals were exposed to Cry1Ab or Cry2Ae proteins in the diet throughout the study, the study cannot be used for avian risk assessment.

	Avian Oral Toxicity  

Rationale was presented to justify waiving the avian oral toxicity data requirement for the Cry1Ab and Cry2Ae proteins expressed in Event T304-40 x Event GHB119 cotton (MRID 47903101). The justification rested primarily on the expected minimal exposure of birds to Cry1Ab and Cry2Ae proteins in TwinLink(TM) cotton as a result of their avoidance of cotton seed consumption, as well as the lack of observed effects of other Bt proteins that have been observed on birds. However, this rationale did not account for other routes of oral exposure in birds (e.g., consumption of insects). BPPD has concluded that currently registered Cry proteins, including Cry1Ab, do not pose hazards to birds (Mendelsohn et al. 2003; U.S. EPA 2005a, 2005b, 2008a, 2008b, 2010). BPPD also subsequently received additional data to show that Cry2Ae protein is not likely to cause adverse effects when raw Event GHB119 cottonseed meal is included at 10% in the diet (MRID 48177501). See the discussion of synergism and protein expression equivalence that supports acceptability of bridging these data to TwinLink(TM). 

		Avian Inhalation Toxicity

Avian inhalation toxicity testing is typically not required for Bt cotton event registrations, since exposure to Cry proteins via inhalation is not expected. The applicant submitted data waiver rationale (MRID 47871603) that is consistent with BPPD's position on inhalation exposure for birds in Bt cotton, and this rationale is acceptable. An avian inhalation test is not required.

   	b. Wild mammals

BCS submitted acceptable data waiver rationale (MRID 47871603) to justify that testing with wild mammals is not necessary. As described above in the risk assessment of Cry1Ab protein in T304-40 and Cry2Ae protein in GHB119 cotton, acute oral toxicity studies with mice are available, and the Agency has no evidence to suggest that laboratory mammals are insufficient surrogates for determining toxicity to wild mammals. See the discussion of synergism and protein expression equivalence below. Protein expression is shown to be equivalent between TwinLink(TM) cotton and the T304-40 and GHB119 parent lines. Consequently, bridging these data to TwinLink(TM) is acceptable. 

		c. Freshwater animals

Data waiver rationale (MRID 47871603) was submitted to fulfill the data requirements for freshwater fish for Cry1Ab and Cry2Ae proteins in TwinLink(TM). Justification rested primarily on the lack of expected exposure of aquatic environments to Cry proteins produced in cotton, and this rationale was determined to be sufficient. BPPD previously determined that exposure to Cry proteins in aquatic environments as a result of registered uses of transgenic cotton is very low (e.g., see US EPA 2008a), and the data waiver rationale was consistent with BPPD's conclusion. 

Toxicity data with Daphnia are available for both Cry1Ab and Cry2Ae proteins for use in the TwinLink(TM) risk assessment. Results of these studies are presented above.   

		d. Estuarine and marine animals

The Agency typically waives toxicity studies with estuarine/marine fish and invertebrates for Bt Cry proteins; however, the applicant submitted data waiver rationale (MRID 47871603) to fulfill data requirements for estuarine and marine animals. The waiver rationale presented was the same as that discussed above for freshwater fish and was determined to be sufficient based on the Agency's conclusions about exposure in aquatic environments. 

   	e. Terrestrial and aquatic plant species

The Agency typically waives nontarget plant testing for Bt Cry proteins, since the active ingredient is an insect toxin (Bt -endotoxin) that has never shown any toxicity to plants. The applicant submitted acceptable data waiver rationale (MRID 47871603) that was consistent with the Agency's position on nontarget plant risks to Bt -endotoxin. 
		
   	f. Invertebrate species        

A 28-day study with collembola (Folsomia candida) was conducted to determine the effects of TwinLink(TM) plant tissue on this soil invertebrate (MRID 47640010). Collembola (Folsomia candida) were provided diet containing nominal concentrations of 5.0, 20, or 50% transgenic lyophilized TwinLink(TM) cotton leaf. The test also included a negative control group of collembola fed diet without lyophilized TwinLink(TM) cotton leaf, a control substance group fed diet containing 50% lyophilized nontransgenic Coker 315 (near isoline) cotton leaf, and a positive control group fed diet containing thiodicarb. Mortality in the 5%, 20%, and 50% TwinLink(TM) test groups was 12.5%, 10.0%, and 7.5%, respectively, while cumulative number of offspring per female were 42.2, 41.7, and 33.4, respectively. Mortality in the negative control and Coker 315 groups was 2.5% and 5.0%, respectively, and cumulative number of offspring per female was 26.9 and 38.6, respectively. Mortality and number of offspring were not statistically significant different among any groups tested. Response in the positive control group was as expected. The results of the study show that exposure to transgenic TwinLink(TM) cotton leaves at 5.0, 20, and 50% concentration in the diet has no significant effect on survival or reproduction of Collembola. 

A study was also conducted to determine the potential dietary effects of TwinLink(TM) pollen when administered to 2-3 day old honey bee larvae (Apis mellifera, MRID 47640009). Honey bee larval cells were treated with 10 uL of a 30% solution containing 2 mg of TwinLink(TM) pollen. Additional treatment groups included: 1) a negative control (pollen from Coker 315) in combination with 30% sucrose; and 2) a positive control (potassium arsenate in 30% sucrose at 2,000 ug/mL). The amount of pollen administered to the larvae was approximately the maximum amount that is typically fed to honey bee larvae (Babendreier et al. 2004). Emergence was complete on Day 15 of the study. Larval survival and adult emergence was determined (96.25% for TwinLink(TM); 97.5% for the Coker 315; and 0% for the reference substance). Based on statistical analysis, there were no significant differences in survival between the TwinLink(TM) treated larvae and those treated with Coker 315 pollen.

Additionally, bridging criteria are satisfied. Consequently, data developed for the Cry1Ab and Cry2Ae proteins expressed individually in Event T304-40 and Event GHB119 cotton plants, respectively, can also be used for the ecological risk assessment for TwinLink(TM).

	g. Supplemental Information

BCS submitted a summary of nontarget organism testing with Cry1Ab and Cry2Ae proteins (MRIDs 47640008), which summarized the toxicity information and data waiver rationale for these proteins and TwinLink(TM) and included an assessment of ecological. This summary was determined to be supplemental, but provided little additional information for use in the risk assessment.   

	2. Synergism of Cry1Ab and Cry2Ae proteins

A sensitive-insect dose-response bioassay with Heliothis virescens neonate larvae was performed to determine whether Cry1Ab and Cry2Ae proteins expressed in TwinLink(TM) cotton plants exhibit synergistic toxicity. Leaf material from the single event cotton plants expressing either Cry1Ab or Cry2Ae protein and the combined event (TwinLink(TM)) cotton plants were fed to the larvae at multiple concentrations determine the LD50s for the proteins within each plant. Expression of each protein was determined to be equivalent between the TwinLink(TM) plants and the individual parent lines. Based on the predicted and actual LD50s for the proteins in each type of plant, a conclusion of no synergism was made. The results of toxicity tests with each individual protein are assumed to adequately represent the toxicity of these proteins when expressed in TwinLink(TM) cotton, based on the equivalent protein expression. 

      3. Protein Expression Comparability

Data have been submitted to the Agency showing the proteins expression of Cry1Ab in Event T304-40 cotton, Cry2Ae in Event GHB119 cotton, and both proteins in TwinLink(TM) cotton. Additional data have been submitted to determine whether the relative potency of plant-produced Cry1Ab and Cry2Ae proteins is similar to the purified bacterially-produced proteins used in the toxicity testing with each individual protein. These data show that each event produced comparable protein expression levels of Cry1Ab and Cry2Ae in each respective parent line as well as the T304-40 x GHB119 hybrid, and that the relative potency of plant-produced and bacteria-produced Cry1Ab and Cry2Ae proteins are equivalent (see Appendices A and B). Consequently, bridging from tests for the individual events is appropriate for support of the T304-40 x GHB119 (TwinLink(TM)) registration. 

      C. CONCLUSION

As long as BCS shows that Cry1Ab and Cry2Ae protein expression is equivalent between TwinLink(TM) cotton and the individual T304-40 and GHB119 parent lines, then no unreasonable adverse effects to nontarget birds, mammals, aquatic animals, terrestrial and aquatic plants, and nontarget invertebrates will occur as a result of exposure to these proteins from registered use of TwinLink(TM) cotton. This conclusion is based on the environmental risk assessments for Cry1Ab and Cry2Ae proteins expressed individually in Event T304-40 and GHB119 cotton, as well as additional data developed with TwinLink(TM) cotton plant material. 

Both Cry1Ab and Cry2Ae proteins have not been shown to have toxic effects on birds, mammals, freshwater invertebrates, terrestrial nontarget invertebrates, and plants, and because exposure is not anticipated in aquatic environments. Bioequivalence of protein expression has been discussed in Appendices A and B. Thus, the Agency makes a "No Effect" determination for direct and indirect effects to listed species of these taxa for these proteins as expressed in TwinLink(TM) cotton. 

E.	References

Babendreier, D., N. Kalberer, J. Romeis, and P. Fluri. 2004. Pollen consumption in honey bee 
      larvae: a step forward in the risk assessment of transgenic plants. Apidologie 35: 293-300.

Marvier, M., McCreedy, C., Regetz, J. & Kareiva, P. 2007. A meta-analysis of effects of Bt 
      cotton and maize on nontarget invertebrates. Science 316: 1475 - 1477.

National Academy of Science (NAS). 2000. Environmental Effects of Transgenic Plants: The Scope and Adequacy of Regulation is available from the National Academy Press, 2101 Constitution Avenue, N.W., Lockbox 285, Washington, DC 20055; (800) 624-6242 or (202) 334-3313 (in the Washington metropolitan area); http://www.nap.edu.

Romeis, J., M. Meissle, and F. Bigler. 2006. Transgenic Crops expressing Bacillus 
      thuringiensis toxins and biological control. Nature Biotechnology 24:  63-71. 

Rosi-Marshall E. J., J. L. Tank, T. V. Royer, M. R. Whiles, M. Evans-White, C. Chambers, N. A. Griffiths, J. Pokelsek, and M. L. Stephen. (2007). Toxins in transgenic crop byproducts may affect headwater stream ecosystems. PNAS:  104(41): 16204 - 16208.

Sanvido,O., Romeis, J., Bigler, F. 2007. Ecological Impacts of Genetically Modified Crops: Ten Years of Field Research and Commercial Cultivation. Adv Biochem Engin/Biotechnol 107: 235 - 278.

Saxena, D. and Stotzky, G. 2001. Bacillus thuringiensis (Bt) toxin released from root exudates and biomass of Bt corn has no apparent effect on earthworms, nematodes, protozoa, bacteria, and fungi in soil. Soil Biol. Biochem. 33: 1225 - 1230. 

U.S. EPA. 1998. "Guidelines for Ecological Risk Assessment."  EPA 630/R-95-002F. Washington, DC, USA. Federal Register, May 14, 1998. 63(93): 26846-26924.

U.S. EPA. 2000. SAP report No 99-06. Sets of scientific issues being considered by the Environmental Protection Agency regarding: Section I - Characterization and Nontarget Organism Data Requirements for Protein Plant Pesticides. Dated February 4, 2000. Available at the EPA website: http://www.epa.gov/scipoly/sap/1999/index.htm#december

U.S. EPA. 2001a. SAP Report No. 2000-07. Sets of scientific issues being considered by the Environmental Protection Agency regarding: Bt plant-pesticides risk and benefit assessments. Dated March 12, 2001. Web site: http://www.epa.gov/scipoly/sap/2000/october/octoberfinal.pdf

U.S. EPA. 2001b. Biopesticides Registration Action Document: Bacillus thuringiensis (Bt) 
      Incorporated Plant Protectants. Washington, DC. http://www.epa.gov/opp00001/biopesticides/pips/bt_brad.htm

U.S. EPA. 2002. SAP Report No. 2002-05. A set of scientific issues being considered by the Environmental Protection Agency regarding: Corn rootworm plant-incorporated protectant nontarget insect and insect resistance management issues. Dated November 6, 2002. http://www.epa.gov/scipoly/sap/2002/august/august2002final.pdf

U.S. EPA. 2004. SAP Report No.2004-05. Product characterization, human health risk, 
      ecological risk, and insect resistance management for Bt cotton products. Dated August 19, 2004. http://www.epa.gov/scipoly/sap/meetings/2004/june/final1a.pdf
      
U.S. EPA. 2005a. Biopesticides Registration Action Document: Bacillus thuringiensis Cry2Ab2 
      Protein and its Genetic Material Necessary for its Production in Cotton. Washington, DC. http://www.epa.gov/pesticides/biopesticides/ingredients/tech_docs/brad_006487.pdf

U.S. EPA. 2005b. Biopesticides Registration Action Document: Bt Cry1F (synpro) and Cry1Ac (synpro) Construct 281/3006 Insecticidal Crystal Proteins as Expressed in Cotton. http://www.epa.gov/oppbppd1/biopesticides/ingredients/tech_docs/brad_006512.pdf
U.S. EPA. 2006. The product characterization and human health data submitted in support of the EUP. Memorandum from R. Edelstein, through J. Kough, to S. Matten dated January 27, 2006. U.S. EPA. Washington DC
U.S. EPA, 2008a. Biopesticides Registration Action Document  -  Bacillus thuringiensis modified Cry1Ab (SYN-IR67B-1) and Vip3Aa19 (SYN-IR102-7) insecticidal proteins and the genetic material necessary for their production in COT102 X COT67B cotton, http://www.epa.gov/pesticides/biopesticides/ingredients/tech_docs/brad_006529.pdf


U.S. EPA, 2008b. Biopesticides Registration Action Document  -  Bacillus thuringiensis Cry2Ab2
      protein and its genetic material necessary for its production in cotton. Amended. 
      http://www.epa.gov/oppbppd1/biopesticides/ingredients/tech_docs/brad_006487.pdf

U.S. EPA, 2010. Biopesticides Registration Action Document  -  Cry1Ab and Cry1F Bacillus thuringiensis (Bt) Corn Plant-Incorporated Protectants, http://www.epa.gov/oppbppd1/biopesticides/pips/cry1f-cry1ab-brad.pdf.

U.S. EPA. 2010. Review of Product Characterization Data in support of Sec. 3 Registration of Plant-Incorporated Protectant (PIP) cotton Events: GHB119 cotton expressing Bacillus thuringiensis Cry2Ae protein (EPA File Symbol 264-RNOU), T304-40 cotton expressing Bt CrylAb (264-RNOL); and TwinLink(TM) cotton (EPA File Symbol EPA File Symbol 264-RNOA), its associated combination PIP product developed by conventional breeding of cotton Events GHB119 and T304-40. Memorandum from A. Waggoner, through J. Kough, to S. Bacchus and D. Greenway dated February 4, 2011. U.S. EPA. Washington DC.

U.S. EPA. 2011. Review of Product Characterization and Human Health Data in support of Sec. 3 Registration of Plant-Incorporated Protectant (PIP) Event T304-40 cotton expressing Bacillus thuringiensis Cry1Ab insecticidal protein (EPA File Symbol 264-RNOU; OECD Unique Identifier: BCS-GHØØ4-7);  Memorandum from A. Waggoner, through J. Kough, to S. Bacchus and D. Greenway dated November 30, 2011. U.S. EPA. Washington DC.
APPENDIX D: Insect Resistance Management

Cry1Ab, Cry2Ae proteins and their combination (Cry1Ab x Cry2Ae) in TwinLink(TM)

Discussions of EPA's risk assessments for Insect Resistance Management (IRM) are presented in this Appendix. Where deficiencies were identified in the earlier reviews, BCS provided data to satisfy EPA's requirements. Immediately below is the risk assessment completed by EPA's J. Martinez on March 22, 2010. Following this risk assessment are those completed on September 28, 2010 and June 28, 2011.
 
I. Insect Resistance Management  Review 

Below is the BPPD review of BCS' Insect Resistance Management (IRM) plan and supporting data for the application to register TwinLink(TM) cotton PIP as well as supporting documentation for a Natural Refuge request (U.S. EPA memorandum 04/22/10; EPA File Symbol 264-RNOU, MRIDs 4760014 and 47634823). Deficiencies identified in this application for full commercial registration TwinLink(TM) cotton were satisfied in subsequent submissions in 2010 and 2011. Those risk assessments follow this initial review.

CONCLUSION AND RECOMMENDATIONS

1. DOSE

   a. BPPD concludes that the diet incorporated and overlay assays together (using verification method 1) suggest that TwinLink(TM) and GHB119 may express a high-dose against CBW but that T304-40 may express less than high-dose against the target pest. 
   
   b. BPPD concludes that diet incorporated and overlay assays (using verification method 1) together suggest that Events GHB119 and T304-40 may express a high-dose and TwinLink(TM) may express an effective high-dose against TBW. 

   c. BPPD cannot make a conclusion about the dose profile for TwinLink(TM) and its events and PBW (using verification methods 3 and 4) because it is unclear how the mortality results reported by Ellsworth and Li (2008) were obtained. When BPPD divided the number of entry holes reported (Tables 9-12) by the number of live larvae and calculated average mortality, the obtained results did not match mortalities reported by BCS. 

BPPD's conclusion regarding dose determination (see also Table D.15)
                                     Pest
                                   Method #1
                                   Method #4
                                   Method #3

                                    T304-40
                                    GHB119
                                      TL
                                    T304-40
                                    GHB119
                                      TL
                                    T304-40
                                    GHB119
                                      TL
CBW
                                 Non-High Dose
                              Possible High-Dose
                         Possible Effective High-Dose
                                   Inconcl.*
                                   Inconcl.*
                                   Inconcl.*
                                      --
                                      --
                                      --
TBW
                              Possible High-dose
                                   High-Dose
                              Effective High-Dose
                                      --
                                      --
                                      --
                                      --
                                      --
                                      --
PBW
                                      --
                                      --
                                      --
                                   Inconcl.*
                                   Inconcl.*
                                   Inconcl.*
                                      ---
                                   Inconcl.*
                                   Inconcl.*
* Insufficient information was submitted to the Agency to make a dose determination 
** Cry1Ab protein could not be tested in this experiment because the lab strain was resistant to Cry1Ab. 


2. PROTEIN EXPRESSION

   a. BPPD has reviewed BCS' protein expression results reported in their IRM submission and concludes that the single Event T304-40 may not express the same levels of toxin as in the dual gene PIP, TwinLink(TM). Event T304-40 expressed higher levels of toxin in leaves of single event plants than leaves collected from TwinLink(TM) plants. For flowers and bolls the pattern was reversed. It is unclear whether these differences in expression results were statistically significant. Because BCS did not provide further information on the differences observed. 

   b. Expression of GHB119 in leaves from early to late stage appeared somewhat variable or unstable. Expression levels were measured in leaves in 2004 and 2005 at three developmental stages (early leaf, mid leaf, and late leaf). At developmental stage 1, expression was higher than at developmental stage 2, and at developmental stage 3, expression increased again but to higher levels than observed at developmental stage 1. For early, mid, and late boll, Expression level for GHB119 were also assessed in three different developmental stages of bolls (early, mid, and late). Levels of GHB119 were higher in TwinLink(TM) than in the single event plant during developmental stage 2 and 3. It appears that the difference was approximately two-fold (see Figure 1). 

   c. Table D.1 in this review summarizes the protein expression level (ug/g-dry weight) submitted by BCS for T304-40 and GHB119 in different plant tissues. It appears that the same tissue types were lumped together for plants expressing single and pyramided events. Since there appears to be unequal expression of the proteins in plants expressing either the single or pyramided events over the course of the growing season, this approach is not providing the most information. It would be more useful to have the expression data separated by tissue type for each individual Event, GHB119, T304-40, and TwinLink(TM), rather than lumped together for the same protein.

   d. BPPD recommends that BCS provide more specific and explanatory information together with their protein expression results in future data submissions for PIP registrations. 




3. CROSS-RESISTANCE

   a. BPPD concludes that based on BCS's data there is little concern for cross-resistance between Cry1Ab and Cry2Ae proteins as expressed in TwinLink(TM).
   
   b. BPPD notes that should resistance evolve in CBW and TBW to Cry2Ae protein (as expressed in TwinLink(TM) cotton) that Cry2Ab2 protein (as in Bollgard II cotton) may be compromised as well because, as BCS stated in their report, there is a great potential for cross-resistance between the two toxins. 

   c. BPPD notes that the cross-resistance potential is great between Cry1Ab, Cry1Ac, and Cry1F (although somewhat less for Cry1F) proteins based on the competition assays; the three toxins share high affinity binding sites. Should CBW evolve resistance to Cry1Ab protein, other PIP products expressing Cry1Ac and Cry1F proteins will likely be compromised as well. 

 4. MODELING

      BPPD has reviewed the modeling assessment for CBW and accepted the probabilistic approach presented therein. However, BPPD is concerned about parameter assumptions for initial resistance allele and dose mortality that went into the analysis, the lack of including Cry1Ab corn in the modeling landscape, the lack of establishing a high dose for TBW (requested for natural refuge strategy request) and the oversight of not modeling PIP durability for the TBW target pest. These concerns are all addressed in the Deficiencies 2-6 listed below.

5. REFUGE STRATEGY

   a. An IRM plan waiver request for single trait Events (T304-40 and GHB119) is acceptable if seeds from those Events are not sold commercially and serve only for producing the combination PIP TwinLink(TM). For any other combinations, BCS may need to apply for an experimental use permit or  product  registration. BPPD recommends the use of acreage/geographic limitations for the single trait Events as a means to mitigate potential resistance development. BCS should provide information on proposed acreage for these single trait Events.

   b. BPPD verified that BCS cited all the relevant natural refuge studies originally developed by Monsanto Company and has obtained a data citation permission letter from Monsanto.

   c. BPPD notes that a second dose verification method is lacking for TBW. Natural Refuge requests were previously granted on the basis that the toxins were expressed at a high-dose against TBW. A second dose TwinLink(TM)-verification method will need to be submitted to support BCS' Natural Refuge request, or BCS can design a model with their specific dose data for TBW.

   d. When BCS has satisfactorily addressed the deficiencies related to the modeling of resistance evolution in CBW and TBW, BPPD will re-assess the natural refuge request for TwinLink(TM) cotton. 

   e. If a Natural Refuge strategy is approved by BPPD for TwinLink(TM), the requirements for a non-Bt cotton refuge in conjunction with the planting of any acreage of TwinLink(TM) cotton should remain in place for the following states and regions: Arizona, California, and New Mexico and in the following Texas counties:  Brewster, Crane, Crockett, Culberson, El Paso, Hudspeth, Jeff Davis, Loving, Pecos, Presidio, Reeves, Terrell, Val Verde, Ward, and Winkler. 

   f. For those States and regions requiring a non-Bt refuge, the following refuge planting requirements should apply:

         i. 5% External, Unsprayed Refuge
         ii. 20% External Sprayed Refuge
         iii. 5% Embedded Refuge 


 6. RESISTANCE MONITORING

   a. BCS' approach for developing baseline data for TBW, CBW, and PBW by 2012 may be too protracted in the case that TwinLink(TM) is commercialized before 2012. Baseline susceptibility data should be available by the first growing season of commercial use of the PIP product. BPPD recommends that BCS accelerate the development of these data. 

   b. The applicant proposed to monitor for resistance evolution in all three target pests (CBW, TBW, and PBW) in areas of high TwinLink(TM) adoption. BPPD agrees with this proposal but also recommends that monitoring efforts additionally focus on regions in states with little natural refuge.

   c. BCS should follow the monitoring paradigm established for other Bt cotton registrations; this point is addressed in Deficiency 8.

7. REMEDIAL ACTION PLAN
   a. BCS needs to have an established remedial action plan in place for each target pest of TwinLink(TM). This is addressed in Deficiency 8 below. 
   b. BPPD notes that it will not be sufficient if BCS "attempts" to confirm and characterize resistance upon receiving notification of unexpected damage. Rather, unexpected pest damage reports should be taken seriously and followed up immediately to conclusively determine whether the cause of crop damage is due to resistance or other reasons. BPPD is aware that it can take months before a determination can be reached.

8. GROWER EDUCATION

   a. It is recommended that BCS develop a grower education program that targets states and counties where a non-Bt corn variety must be planted as the refuge for TwinLink(TM) cotton. 
   
   b. It is recommended that BCS develop a grower guide that at a minimum describes the technology and includes unexpected damage guidelines.

      
9. GROWER COMPLIANCE

   a. BPPD notes that BCS should have a well developed grower-education program (including grower guide) for all regions with clear indications for regional requirements.

   b. A natural refuge strategy for TBW and CBW would eliminate the need for growers to plant a refuge where alternate hosts are abundant. However, there are cotton-growing regions where natural refuge is not applicable and where planting a refuge is still required. For those regions it is recommended that BCS develop, implement, and report to EPA on programs to evaluate and promote growers' compliance with IRM requirements. 

   c. Grower compliance in these regions should be assessed with on-farm assessments in addition to anonymous phone surveys. BPPD recommends that BCS use the existing grower compliance paradigm for other Bt cotton registrations.

10. REPORTING REQUIREMENTS

      BCS stated that "once the baseline susceptibility data were collected for the three primary target pests and a discriminating dose level was established, annual monitoring would be conducted and reported to the agency". BPPD recommends that BCS submit annual reports for the following:
         i) Baseline and discriminating dose work irrespective of their status;
         ii) Sales data;
         iii) Annual resistance monitoring; and 
         iv) CAP results for cotton growing regions where Natural Refuge is not applicable.
11. DEFICIENCIES

Deficiency 1: The dose profile reported for PBW cannot be verified by BPPD until BCS clarifies how mortalities were calculated for the experiments using high-dose verification methods (3) and (4).

Deficiency 2: In the modeling analysis provided by BCS, only CBW was modeled but not TBW. For a new refuge strategy such as Natural Refuge, each main target pest needs to be evaluated separately via a simulation model that incorporates biology, genetics, toxicity, and ecology of the target pest. BCS needs to provide simulation modeling for TBW in order to support their natural refuge request. 

Deficiency 3: The natural refuge paradigm was approved for Bollgard II and WideStrike with a high-dose profile for TBW. If BCS wishes to pursue the natural refuge strategy for TwinLink(TM), then a second high-dose verification study will need to be submitted to support that the PIP expresses a high-dose against TBW, or BCS can design a model with their specific dose data for TBW.

Deficiency 4: BPPD is concerned with the dose assumptions for CBW and recommends that an additional analysis be conducted using somewhat lower mortality assumptions based on the study results reported by BCS. The bioassay experiments showed that Cry1Ab protein caused mortality ranging 88.5-100% (see this review). Estimated mortalities from the artificial infestation experiments (2007-2008) ranged from 99.5-99.7% (see this review). However, BPPD cannot verify the results of the artificial infestation experiments due to lack of information provided (i.e. % egg viability). At the moment, there is no evidence that the Cry1Ab PIP expresses high-dose against CBW. BCS concluded, too, that TwinLink(TM) Cotton expressed less than high-dose against CBW due to the target pest being less susceptible to Cry1Ab and lower level of expression of Cry1Ab toxin in the parental Event for TwinLink(TM) Cotton. Hence BPPD recommends exploring dose mortalities ranging from 90-99% in the simulation model.

Deficiency 5: BPPD is concerned about the most likely value chosen for initial resistance allele frequency for Cry1Ab (0.0002). Although no field resistance to Cry1Ab has been documented yet for corn earworm (aka CBW), the toxin has been in use for over a decade (in Bt corn) and selection may likely have increased the frequency of potential resistance alleles. However, because corn earworm does not overwinter in the Corn Belt, BPPD agrees that the most likely value will be somewhat lower than could otherwise be expected. BPPD recommends that BCS consider modeling a most likely resistance allele frequency of 0.001 for Cry1Ab in CBW.

Deficiency 6: BCS needs to include Cry1Ab corn and Cry1Ab cotton in the modeling landscape when assessing the risk of resistance evolution in the target pests because Syngenta has registered Bt cotton and Bt corn and Monsanto has registered Bt corn products expressing Cry1Ab. These corn and cotton PIPs are grown in the Cotton Belt and need to be included in a simulation analysis for TwinLink(TM).

Deficiency 7: BCS needs to have a well-developed monitoring plan in place before TwinLink(TM) is approved for commercialization (i.e. sampling regions, sampling scheme, etc). BPPD recommends that BCS follow the monitoring paradigm established for other Bt cotton registrations.

Deficiency 8: BCS needs to have an established remedial action plan in place for each target pest of TwinLink(TM). It is not sufficient to provide an outline or general statement about different kinds of approaches. BPPD recommends that BCS follow the existing remedial action paradigm for Bt cotton PIPs. 


     I. BACKGROUND  -  Insect Resistance Management

TwinLink(TM) cotton is a pyramided transgenic cotton trait that expresses Cry1Ab (Event T304-40), Cry2Ae (Event GHB119), and a glufosinate ammonium (herbicide) tolerant active ingredient based on LibertyLink technology. TwinLink(TM) was developed by BCS using conventional breeding techniques. 

Cry2Ae is a new PIP for Bt cotton that allows for the control of Lepidopteran pests. Small field trial evaluations with artificially and naturally infested plots using the three major target pests Cotton bollworm (Helicoverpa zea, CBW), Tobacco budworm (Heliothis virescens, TBW), and Pink Bollworm (Pectinophora gossypiella, PBW) demonstrated that there was minimal damage to TwinLink(TM) Cotton (expressing Cry2Ae and Cry1Ab proteins) during the growing seasons of 2004 and 2005.

Cry1 and Cry2 are effective lepidopteran toxins and share less than 20% amino acid homology. Crickmore et al. (1998) constructed a phylogenetic tree using amino acid sequence identity as homology criteria for all presently known Bt crystal proteins. This tree indicates that Cry1Ab and Cry2Ae proteins are genetically dissimilar entities and share little sequence homology. 

Currently there are five other Cry toxins registered in Bt cotton for the control of Lepidopteran pests; those are Cry1F, Cry1Ab, Cry1Ac, Cry2Ab2, and Vip3Aa19. Pyramids of Cry1F and Cry1Ac, Cry1Ac and Cry2Ab2, and Cry1Ab and Vip3A19 are expressed in WideStrike cotton (product of Dow AgroSciences), Bollgard II cotton (product of Monsanto), and VipCot (product of Syngenta), respectively. All pyramided cotton PIP are effective against the three main target pests CBW, TBW, and PBW. 

II.       PEST BIOLOGY AND ECOLOGY

TBW, CBW, and PBW are considered the three major target pests for TwinLink(TM) Cotton. Fall armyworm (Spodoptera frugiperda, FAW) and Beet armyworm (Spodoptera exigua, BAW) are considered secondary target pests. A summary of the biology and ecology for major Bt cotton target pest, such as TBW, CBW, and PBW, can be viewed in the IRM section of the Agency's 2001 Bt crop reassessment document at http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad.htm. Since the publication of the Agency document, additional biological and ecological information has become available. BCS summarized new study results (MRID 47640014, Vol. 15), which are reported below.

                     1.      Cotton Bollworm (Taken from MRID 47640014)
     
CBW is highly mobile and can move long distances. The general pattern of CBW migration is a northward movement (U.S. EPA, 2001). At the 2001 Scientific Advisory Panel (SAP) the general consensus was that southward movement to overwinter was not proven but that there was good circumstantial evidence. Further research on south migration was done by Gould et al. (2002): they used 13C/12C stable isotope assessment of alternative host used by CBW and showed indirectly that late season CBW moths captured in Louisiana and Texas could be migrants from the northern Corn Belt. Typically, greater than 50% of the moths captured in August through October have isotope ratios indicative of larval feeding on C4 plants (i.e. tropical and prairie grassland plants), even though local C4 crops were harvested at least one month earlier and the authors presume at the Texas trial site that there are also no common wild C4 plants. Cotton, along with wheat, rice, barley and other ancient plants are C3 plants, easily differentiated from C4 plants.

Directly related to migration is the overwintering behavior of the insect. Schneider (2003) studied the overwintering of CBW in cotton fields in the northeast portion of Mississippi. He found that densities of overwintering pupae varied greatly across years; in non-Bt plots densities ranged from 0-67 pupae/ha for CBW. Lower densities of CBW pupae (0-214 pupae/ha) were found over the seven-year study in Bt cotton fields. Only partial control of CBW was possible with these first generation single gene Bt cotton plants expressing only the Cry1Ac protein.

Local temporal and spatial dispersal in CBW (Han and Caprio, 2002) were studied by using polymorphic allozyme and RAPID-PCR makers. For CBW, Han and Caprio (2002) found that population differentiation at local and regional areas was very low, not unexpected for a highly polyphagous and migratory insect. Contrary to TBW, the genetic variance was not significantly different between the four seasonal generations for CBW. 

Similarities amongst pheromone trap catches of CBW in eastern North Carolina suggested a high level of CBW moth movement into and out of localized areas and was much more significant than anticipated (Jackson et al., 2003). Cotton production in localized areas has little impact on the number of CBW moths in the local environment. 

Gore et al. (2002) monitored larval CBW movement on cotton and compared movement on Bt cotton (Bollgard) and non-transgenic cotton. On pre-flowering plants at 1, 3, 6, and 24 hours after infestation 47.8%, 39.4%, 20.9%, and 13.0% of the larvae, respectively, remained in the terminals of non-Bt cotton plants. For Bt cotton, this was 28.7%, 11.4%, 6.3%, and 1.3%, respectively. A day after artificial infestation, larvae were found 2.4 main stem nodes below the terminals on flowering non-Bt plants, for Bt cotton this was 5.7 main stem nodes below the terminals. 

In the Mississippi Delta, moth populations were monitored from 1986 to 2005 (Adamczyk and Hubbard, 2006) with pheromone traps. During the non-crop months, moderate numbers of moths of CBW were typically captured by 20 March. From 1986-1996, the mean number of CBW moths captured per trap and per night was 15.7 with a maximum of 310. Since 1997, CBW populations have declined, with a mean number per trap and per night of 6.2, and a maximum of 70. The decline may due to wide-scale planting of Bt cotton.

Jackson et al. (2004) estimated the number of CBW that developed on conventional and Bt cotton in 2000-2002 for North Carolina. Across 7 locations they found 60,283 4[th] and 5[th] instar living larvae/ha in conventional cotton, which produced 47,078 adults/ha and damaged 367,835 bolls/ha. The Bt cotton (Bollgard event) did not significantly reduce larval numbers, but damaged bolls and adults were significantly less. The numbers of larvae were also estimated in fields utilizing the conventional chemical insecticide regime in 2001-2002; non-treated fields produced 75,510 larvae/ha, while sprayed fields had significantly lower levels, only 28,973 larvae/ha. 

Alternative cultivated or wild hosts, which can be used as non-crop or natural refuges, have been extensively studied since 2001. Jackson et al. (2003) studied the larvae, pupae and adult populations on different hosts in eastern North Carolina and found that corn was the major producer of CBW larvae in the early season and stayed a major producer until the last week of August. These results indicated that field corn may serve as refuge for CBW much longer than originally expected. Beginning in mid-August, soybean and conventional cotton became the major crop sources for CBW production. Pupae and adult production followed the same trends. 

Kurtz et al. (2004) used these estimates (Jackson et al., 2003) and unpublished data by Jackson in their computer simulation model. For the first two CBW generations, 90% of the larvae developed on corn, 5% on cotton and 5% on soybean. For the third CBW generation in eastern North Carolina the estimates dropped to only 17.39% production on corn, with increases to 36.34% production on cotton, 0.05% on peanuts and 40.99% on soybean. Storer et al. (2001) confirmed these results, finding that CBW adults emerging from ear-stage cornfields infest cotton, soybean and other crops In North Carolina. The level of adult CBW production was greatly decreased (~80%) in Bt corn, while non-Bt corn produced 20-30 adults per 100 ears.

In Louisiana, Peters et al. (2004) monitored larval CBW populations in several field crops during the 2002 and 2003 growing seasons. Larval populations were highest in sorghum: approximately 32,000 larvae/acre in 2002 and approximately 12,000 larvae/acre in 2003; field corn: approximately 20,000 larvae/acre in 2002 and approximately10,000 larvae/acre in 2003. Larval densities on soybean and on non-transgenic and Bt cotton were much lower (<3000 larvae/acre). Larval populations in corn, sorghum, maturity group 6 soybean, cotton and Bt cotton peaked on 25 June, 12-14 July, 2-4 August, 2-4 August, and 11-16 August, respectively. The only crop that exhibited similar temporal synchrony with bollworm larvae on Bt cotton was soybean maturity group 6, indicating the possibility of including it in an IRM strategy. A more extensive study over two years and over five states demonstrated again that non-cotton crops, such as corn, grain sorghum, peanut and soybean, support much larger larval CBW populations than cotton throughout the season. In June and portions of July the larval CBW populations were almost entirely restricted to corn and in July and August were observed in non-cotton crops more frequently and in larger numbers than in cotton. Again it was shown that CBW moths move extensively from their natal host origins. The number of CBW adults captured in pheromone traps at interfaces of fields of Bt cotton and various crop hosts rarely varied among interfaces, except for highly attractive corn (Jackson et al., 2008). Tillman and Mullinix (2004) showed that CBW females preferred to oviposit in grain sorghum over cotton; the number of eggs per plant and percentage of plants with CBW eggs was higher in sorghum trap crop than in cotton trap crop for two years.

Larval development studies in the lab of CBW on sorghum, corn, soybean and cotton showed a survival of 73, 55, 26, and 13%, respectively and a mean development time to pupation of 15.6, 12.4, 18.4, and 25 days, respectively (Gore et al., 2003). Cotton was the least suitable host, which was also reflected in the significantly lower pupal weight. Blanco et al. (2007) showed that over a three year period in Mississippi and Mexico, garbanzo bean, an important legume crop in Mexico, had an annual CBW level of 96,000 larvae/ha on garbanzo bean, and approximately 5 times less larvae on cotton and approximately 6 times less larvae on velvetleaf, respectively. Low densities of adult CBW moths (0-275 ha/yr) emerged from plots and were not significantly different between the three crops. Cover crops, such as legume mixture, crimson and rye seem to be more preferred by heliothines than the cotton crop. In sweeps no heliothine could be caught on the cotton plants; in contrast, around 0.5 heliothine larva/sweep were found on the cover crops (Tillman et al., 2004). 

Natural mortality of CBW was studied from 1991-95 in short season cotton in Texas with one generation per year (Sansone and Smith, 2001). Mortality of 71-95% could be consistently expected in the egg and first-instar stages. The more detailed studies in 1991-1993 partitioned the mortality; 3.4% was caused by Trichogramma egg parasitation, larval parasitation rarely exceeded 5%, but numerically the most import mortality factor was egg consumption of around 77% by Orius insidiosus (minute pirate bug). 

                     2.      Tobacco Budworm (Taken from MRID 47640014)

TBW is a highly mobile insect that becomes less mobile in the mid to late summer when it prefers cotton as a host plant. Directly related to movement is the overwintering of the insect. Recently, Schneider (2003) studied the overwintering of TBW in cotton fields in the northeast portion of Mississippi. He found that densities of overwintering pupae varied greatly across years: in non-Bt plots densities ranged from 0-303 pupae/ha TBW. No TBW pupae were found over the seven-year study in Bt cotton fields. Using pheromone traps to estimate the total area-wide TBW population overwintering, Schneider (2003) found that in the pre-transgenic cotton era (<1996) cotton fields typically accounted for up to 10% of the overwintering TBW population. After introduction of Bt cotton, but prior to expanded no-till cotton planting in the 1996-2001 era, the estimate was less than 2%. 

Local temporal and spatial dispersal of TBW (Han and Caprio, 2004) was studied by using polymorphic allozyme and RAPID-PCR makers. TBW populations overall showed little special differentiation, indicating a highly effective gene flow rate over the spatial scale under investigation (Han and Caprio, 2004). Allelic differentiation arose more frequently in the second generation than in the other three generations, suggesting that TBW moth dispersal was more limited during the midsummer. During midsummer, cotton is the main host available for TBW in the Mississippi Delta. Once cotton flowers, moth movement declines along with the availability of a ready nectar source. Because the populations never reached genetic equilibrium, Han and Caprio (2004) could not determine the extent of movement reduction.

In the Texas High Plains TBW adult populations were monitored weekly from March to November between 1982 and 1995 using pheromone traps at Lubbock and Halfway Texas, prior to the introduction of Bt cotton (Parajulee et al., 2004). The bollworm/budworm complex consisted of 98% CBW and 2% TBW. For weekly moth abundance, a positive correlation was shown with weekly temperature, degree-days and weekly average precipitation, and a negative correlation was shown with the weekly wind velocity. Data from this study indicate that the late-season moth catches could be indicative of the dynamics of the early-season catches the following year. 

In the Mississippi Delta, moth populations were monitored from 1986 to 2005 (Adamczyk and Hubbard, 2006) with pheromone traps. During the non-crop months, moderate numbers of moths of TBW were typically captured by April 10[th]. From 1986-1996, the mean number of TBW moths captured per trap and per night was 19.5 with a maximum of 500. Since 1997, populations have declined for TBW with a mean number per trap and per night of 2.9, and a maximum of 25. The decline may be due to wide scale planting of Bt cotton.

In the Southeastern US, the first TBW generation develops almost exclusively on weeds; throughout the whole growing season cotton is a less important host, because of the abundance of other host plants (Abney et al., 2004). Abney et al. (2005) also showed, using 13C/12C stable isotope assessment, that TBW fed consistently on C3 plants (examples of C3 plants are cotton, soybean, clover, and peanut). Unfortunately this technique could not be used to determine the natal origin of the TBW.

Blanco et al., (2007) showed that over a three year period in Mississippi and Mexico, garbanzo bean, an important legume crop in Mexico, had an annual TBW larval density of 287,000 larvae/ha, approximately 4 times higher than in cotton, and approximately 5 times higher than in velvetleaf. Also the TBW moth densities emerging from garbanzo bean (20,180 adults/ha) were approximately 13 times higher than from cotton and approximately 35 times higher than from velvetleaf.

Cover crops, such as legume mixtures, crimson, and rye, seem to be more preferable to heliothines than cotton. In sweeps, no heliothine could be caught on the cotton plants; in contrast, around 0.5 heliothine larva/sweep were found on the cover crops (Tillman et al., 2004). 

                     3.      Pink Bollworm (Excerpted from MRID 47640014, modified by BPPD)
     
Flight-mill experiments by Wu et al. (2006) showed that over a 72h flight the mean accumulated flight distance and flight duration of PBW were 41.25 km and 23.87 h for females and 23.46 km and 14.12 h for males. For unmated moths, the flight activity reached a peak after 3-5 days after eclosion, and then reduced gradually. Optimum temperature for flight was from 24 to 28°C and the optimum relative humidity ranged from 75 to 90%.

While PBW can be good fliers, local adult movement is limited; in most cases dispersal is limited to <1 km (Carrière et al., 2001a). Besides adult movement, oviposition preference is also important. Liu et al. (2002) found that in greenhouse experiments females did not lay fewer eggs on Bt cotton bolls than on non-Bt cotton bolls. They also revealed that the number of eggs laid per boll was negatively associated with boll age and positively associated with boll diameter; more eggs per boll were laid on plants with more bolls. The distribution of eggs was clumped, which is an indication that boll quality rather than avoidance of other eggs was the primary factor in oviposition preference. Liu et al. (2002) also studied mining behavior of neonate larvae. The number of entrance holes per boll did not differ between Bt cotton and non-Bt cotton. As with female moths, neonate preferred younger bolls and larger bolls.

Carrière et al. (2001c) studied the prediction of spring moth emergence and its use for cultural control of the insect. Accumulation of heat units from January 1st accurately predicted the rate of PBW emergence in different regions in Arizona. Emergence was significantly higher in higher altitude regions. Most moths emerge from diapause too early to reproduce on cotton, a phenomenon known as `suicidal emergence'. Manipulation of the planting date could be a good cultural control, but shortening the growing season can reduce yield.

Most larval stages of PBW are relatively protected from natural enemies because they feed on seeds inside the cotton boll; eggs laid externally on vegetative structures, however, are vulnerable to predation and parasitism. Predation rates on sentinel PBW eggs were compared on Bt and non-Bt cotton plants (Sisterson et al., 2004). Many eggs were attacked by predators with pierce and/or sucking mouthparts and by lacewing larvae. Between Bt and non-Bt plots, no significant differences were found for PBW egg predators; approximately 1.5-3 predators per plant were recovered, and PBW egg predation rates were approximately 40-50%. 

In Naranjo (2005), sentinel eggs of PBW also experienced the same rate of predation in Bt and non-Bt cotton. Egg predation rate varied from around 53% in 2001 to approximately 35% in 2003. No parasitism was detected in eggs that survived predation. On average, chewing and sucking predators were each responsible for approximately 50% of the total mortality. Naranjo (2005) also found no significant difference for rates of pupal mortality between Bt and non-Bt cotton. Pupal mortalities of around 40% in 2002 and 64% in 2003 were observed. No parasitism was observed in intact pupae. 




III.       DOSE

TwinLink(TM) cotton is a combined trait product made through traditional breeding of Events T304-40 and GHB119 which express Cry1Ab and Cry2Ae proteins, respectively. BCS submitted 1) protein expression data for different cotton tissues; 2) LC50 data for Cry1Ab and Cry2Ae proteins and their target pests; and 3) dose information for Event T304-40 (Cry1Ab), Event GHB119 (Cry2Ae), and TwinLink(TM) Cotton using methods #1 and #4 as described by the SAP (U.S. EPA, 2001). 
   
   a) Protein Expression
   
BCS submitted protein expression results for cotton tissues, which were measured using ELISA based assays in 2004 and 2005. Bayer determined that protein expression in Cry1Ab and Cry2Ae cotton as well as TwinLink(TM) cotton plants depended on the type as well as maturity of tissues assayed (see Figure 1). The lowest expression was found in nectar, pollen, and seeds; the highest expression was observed in leaves, roots, squares (flower buds containing male and female reproductive parts), apex, stems, and bolls (fertilized ovum followed by enlargement, filling, and maturation of structure) (see Table 1). Cry1Ab protein, as expressed in TwinLink(TM), had consistent expression in leaves and squares with higher expression in bolls and flowers. Cry2Ae protein, as expressed in TwinLink(TM) had relatively high expression levels in leaves across the growing season and in late bolls (data from 2005). Expression was typically higher in Cry2Ae than Cry1Ab PIP (up to 10X). Cotton plants expressing the single events were analyzed in 2004 and showed similar protein expression levels as the plants expressing both events. BCS concluded that there was no suppression or enhancement of protein expression when Cry1Ab and Cry2Ae were expressed in the same plant.



Figure D.1. Expression of Cry1Ab and Cry2Ae as % Bt protein/total extractable protein - TL denotes TwinLink(TM) (extracted from MRID 47640014)




Table D.1. Expression of Cry1Ab and Cry2Ae proteins in plant tissues
                                 PLANT TISSUE
                   AVE EXPRESSION LEVEL (ug/g - dry weight)
                                       
                                    Cry1Ab
                                    Cry2Ae
Roots
                                   2.9-11.2
                                   8.8-19.7
Stems
                                    0.5-7.3
                                    4.4-7.3
Leaves
                                    0.4-5.2
                                   10.7-63.3
Squares
                                    0.8-3.9
                                   3.4-12.4
Apex
                                    1.0-1.6
                                   6.1-10.7
Bolls
                                    0.2-0.6
                                    3.3-4.9
Flowers
                                   8.7-12.5
                                    4.8-6.3
Seed
                                    0.1-0.7
                                   0.02-1.6
Pollen
                                   0.03-0.07
                                   0.06-0.6
Nectar
                                 0.00001-0.005
                                       0
(Extracted from MRID 47640014)


   b) LC50 Data for Cry1Ab and Cry2Ae proteins in CBW and TBW
BCS provided results from insect bioassays using purified individual Bt proteins in a surface layer diet for the target pests of TwinLink(TM) cotton. Table 2 shows the mean LC50 values and ranges for six target pests. BCS described the insect susceptibility for Cry1Ab protein as follows: TBW > PBW > Old World BW (OWBW, Helicoverpa amigera) >> CBW. The susceptibility for Cry2Ae protein was described as: TBW > PBW >> CBW > BAW > FAW > Old World BW.


Table D. 2. Mean LC50 values and ranges for Cry1Ab and Cry2Ae proteins
INSECTS
                                    Cry1Ab 
                                  (ng/cm[2])
                                    Cry2Ae 
                                  (ng/cm[2])
Heliothis virescens, TBW
                                      7.2
                                  (4.9-10.5)
                                     4.1 
                                   (2.9-5.5)
Helicoverpa zea, CBW
                                     1357
                                  (462-2844)
                                    238.9 
                                 (107.7-473.1)
Helicoverpa armigera, Old World BW
                                     78.5
                                 (41.1-149.9)
                                     3556
                                  (2824-4709)
Pectinophora gossypiella, PBW
                                     41.2
                                  (6.9-152.0)
                                     11.6
                                  (4.9-22.2)
Spodoptera frugiperda, FAW
                                 >>2000
                                    1154.3
                                (682.0-1680.1)
Spodoptera exigua, BAW
                                 >>1350
                                     486.4
                                 (336.5-685.5)
 (Extracted from MRID 47640014)


   c) Dose Determination
   
The 1998 SAP Subpanel discussed ways to define and measure "high dose" in plants. It was agreed that the definition of high dose as "25 times the toxin concentration needed to kill susceptible larvae" was reasonable based on current empirical data. However, the Subpanel recognized that it was conceivable that a heterozygote might develop with higher than 25-fold resistance. The major problem identified by the Subpanel was in determining if this 25-fold level was achieved in a specified cultivar. After much discussion, it was concluded that there were at least five ways to assess this 25-fold level, but that some approaches were more appropriate for specific crop pests. The Subpanel concluded that a cultivar could be considered to provide a high dose if two of the five approaches described here indicated presence of a high dose. The five approaches generally used for Lepidopteran systems are:

(1) Serial dilution bioassay with artificial diet containing lyophilized tissues of Bt plants (tissue from non-Bt plants serving as controls);

(2) Bioassays using plant lines with expression levels approximately 25-fold lower than the commercial cultivar (determined by quantitative ELISA or some more reliable technique);

(3) Survey large numbers of commercial plants on sentinel plots in the field (e.g., sentinel sweet corn method) to make sure that the cultivar is at the LD99.99 or higher to assure that 95% of heterozygotes would probably be killed. With this approach Bt sweet corn hybrids are used to attract high densities of ECB and cotton bollworm (Helicoverpa zea)(Boddie)) (CBW/CEW)  moths, sampling can be limited to sweet corn ears in the Bt plot (ca. 1/4-1/2 acre block), and a frequency of resistant phenotypes can be estimated as the ratio of density of larvae/plant in Bt sweet corn to density of larvae/plant in an adjacent planting of non-Bt sweet corn (Andow and  Hutchison, 1998; Hutchison, unpublished data);

(4) Similar to (3) above, but would use controlled infestation with a laboratory strain of the pest that had an LD50 value similar to field strains; and

(5) Determine if an older instar of the targeted pest could be found with an LD50 that was about 25-fold higher than that of the neonate larvae. If so, that stage could be tested on the crop plants to determine if 95% or more of the older stage larvae were killed.

BCS applied three methods to test for high-dose of TwinLink(TM) and its single events. 

   * Method (1) utilized lyophilized plant tissue from Events T304-40, GHB119, and TwinLink(TM) mixed with artificial diet to produce 10X and 25X concentrations. The bioassays were conducted as diet incorporated assays and diet overlay assays with TBW and CBW. 
   * Method (4) involved controlled infestations of TwinLink(TM) cotton plots with susceptible lab strain of CBW and susceptible lab strain of PBW to assess the level of mortality. 
   * Method (3) utilized a homozygous resistant lab strain (Cry1Ac) of PBW with which young cotton bolls were infested.

High-Dose Verification Method (1):

   1. Cotton Bollworm

Lyophilized plant material from greenhouse grown cotton was mixed into the bioassay diets at different concentrations (4% and 10% of Cry-protein). For the negative control, 0, 0.8, 4, and 10% of non-Bt cotton (Coker312) were used. Twenty-four neonate larvae were exposed at each treatment level (see Table D.3); two replicates per treatment were carried out. Mortality was scored at days 4, 7, 12, 17, 19, 21, and 24, and average mortality was calculated from the two replicates.

For overlay assays, the same leaf material was used and the lyophilized powder was mixed in a 0.2% agar solution. The concentrations tested were 4% and 6.9% of Cry-protein. Liquid diet was poured into each well and allowed to dry. Different concentrations were prepared (see Table D.4) and overlaid onto the solidified diet. Mortality was scored at day 3, 7, 11, 15, 18, and 21. 

The applicant reported that control mortality was difficult to manage using cotton leaf tissue due to the presence of gossypol. Diet incorporated assays had higher control mortality (17.5-29.5% on day 17) than overlay assays (16.8-19% on day 15); hence overlay assays were best for verifying TwinLink(TM) cotton's high-dose level. BCS concluded that TwinLink(TM) expressed less than high-dose against CBW due to the target pest being less susceptible to, and lower level of expression of, Cry1Ab protein in the parental event for TwinLink.


Table D.3. CBW mortality to Cry1Ab, Cry2Ae, and TwinLink(TM) Cotton - diet incorporated assays
Treatment
                           % Treatment Concentration
                                  % Mortality

                                       
                                     Day 4
                                     Day 7
                                    Day 12
                                    Day 17
                                    Day 19
                                    Day 21
                                    Day 24
Coker312
                                      10
                                       0
                                     10.5
                                      17
                                     29.5
                            All pupae or pre-pupae

                                       4
                                       4
                                     10.8
                                      15
                                      23
                            All pupae or pre-pupae

                                      0.8
                                       0
                                      8.5
                                     15.5
                                     17.5
                            All pupae or pre-pupae

                                       
                                       
                                       
                                       
                                       
                                       

                                       
                                       
                                       
                                       
                                       
                                       
T304-40, Cry1Ab
                                      10
                                      21
                                     85.5
                                      100
                                       
                                       
                                       
                                       

                                       4
                                       6
                                      21
                                      41
                                      55
                                     57.5
                                     81.5
                                      89
GHB119, Cry2Ae
                                      10
                                      29
                                      72
                                      100
                                       
                                       
                                       
                                       

                                       4
                                      19
                                      31
                                      52
                                     73.5
                                     73.5
                                      84
                                      95
TwinLink(TM)
                                      10
                                      44
                                      85
                                      100
                                       
                                       
                                       
                                       

                                       4
                                     14.8
                                     55.5
                                      70
                                      94
                                      97
                                      100
                                       
Modified from MRID 47640014


Table D.4. CBW mortality to Cry1Ab, Cry2Ae, and TwinLink(TM) - overlay assay method
Treatment
                           % Treatment Concentration
                                  % Mortality

                                       
                                     Day 3
                                     Day 7
                                    Day 11
                                    Day 15
                                    Day 18
                                    Day 21
Coker312
                                      6.9
                                       0
                                     10.5
                                     16.8
                                     16.8
                            All pupae or pre-pupae

                                       4
                                       0
                                       8
                                      17
                                      19
                            All pupae or pre-pupae
T304-40, Cry1Ab
                                      6.9
                                       0
                                     35.5
                                      54
                                     88.5
                                     88.5
                                     88.5

                                       4
                                       0
                                      23
                                     47.5
                                      76
                                      84
                                     89.5
GHB119, Cry2Ae
                                      6.9
                                      23
                                      79
                                      100
                                       
                                       
                                       

                                       4
                                     49.5
                                      83
                                     97.5
                                      100
                                       
                                       
TwinLink(TM)
                                      6.9
                                     27.5
                                      75
                                      100
                                       
                                       
                                       

                                       4
                                      29
                                      80
                                     97.5
                                      100
                                       
                                       
Modified from MRID 47640014


   2. Tobacco Budworm

As for TBW, two types of assays were carried out, the diet bioassay and overlay bioassay using all three toxins. The same methodology was used for TBW as was used for CBW (see above Section III c) 1.). However, concentrations for the negative control of diet bioassays differed and were 0, 0.8, and 4% of Coker312. Furthermore, mortality was scored at day 4, 7, 12, and 17 and averaged across the two replicates. For overlay assays, mortality was scored at day 3, 7, 11, and 15.

As for CBW bioassays, control mortality for TBW was high and difficult to control. The applicant attributed this to the presence of gossypol in the leaf diet. BCS reported that control mortality was lower in overlay assays and that therefore those results were best for verifying TwinLink(TM) Cotton's dose profile. BCS concludes from the bioassay results using method (1) that there was a high dose expression for TBW in TwinLink(TM) cotton and its individual traits.

Table D.5. TBW mortality to Cry1Ab, Cry2Ae, and TwinLink(TM) - diet incorporated assays
Treatment
                           % Treatment Concentration
                                  % Mortality

                                       
                                     Day 4
                                     Day 7
                                    Day 12
                                    Day 17
Food*
                                       0
                                       0
                                      8.3
                                      25
                                      25
Coker312
                                       4
                                       0
                                     14.8
                                     23.8
                                     23.8

                                      0.8
                                       2
                                      17
                                      28
                                      28
T304-40, Cry1Ab
                                       4
                                      29
                                      73
                                     95.5
                                      100
GHB119, Cry2Ae
                                       4
                                      100
                                       
                                       
                                       
TwinLink(TM)
                                       4
                                      100
                                       
                                       
                                       
Modified from MRID 47640014
* Unclear what source of food constituted negative treatment.


Table D.6. TBW mortality to Cry1Ab, Cry2Ae, and TwinLink(TM) - overlay assay method  
Treatment
                           % Treatment Concentration
                                  % Mortality

                                       
                                     Day 3
                                     Day 7
                                    Day 11
                                    Day 15
Coker312
                                       4
                                       0
                                     14.8
                                      17
                                      17
T304-40, Cry1Ab
                                       4
                                      41
                                     88.5
                                     95.5
                                      100
GHB119, Cry2Ae
                                       4
                                      76
                                      100
                                       
                                       
TwinLink(TM)
                                       4
                                      71
                                      100
                                       
                                       
Modified from MRID 47640014


High-Dose Verification Method (4):

   1. Cotton Bollworm
   
Coker312, TwinLink(TM), and its two single events were evaluated for efficacy against CBW and PBW. The experiment was carried out at 14 different locations across the US with two replicates each during the 2007 and 2008 cotton growing season. At two locations only, BCS conducted a larval survival experiment which provided the data for the reported results of high-dose verification method (4). 

In two locations (in 2007) and three locations (in 2008) of the Cotton Belt (actual locations were not reported by the applicant), natural infestation was supplemented with artificial infestations of the target pest; repeated artificial infestations with CBW were conducted at one location in 2007 and three locations in 2008. A total of 7,200 eggs were applied in each row of 60 plants and 50 plants in 2007 and were analyzed for presence of larvae. In 2008, a total of 6,600 eggs/plot were applied and each boll was dissected subsequently to determine presence of larvae. Percent mortality was calculated for 2007 (Table D.7). The average number of boll examined in each plot was recorded and percent mortality calculated in 2008 (Table D.8).

BCS reported that mortality for all three events (single and dual) was very high and greater than 99.2% in all but one instance. In 2008, one TwinLink(TM) plot had slightly less CBW mortality (98.5%).

Table D.7. CBW mortality on artificially infested plots during the 2007 growing season
Event
                               Total eggs/plot*
                               Total live larvae
                                  % Mortality
T304-40, Cry1Ab
                                     6000
                                      16
                                     99.73
GHB119, Cry2Ae
                                     6000
                                       1
                                     99.98
TwinLink(TM)
                                     6000
                                       2
                                     99.97
* 50 plants/60-plant row were scouted = 83.3%
Modified from MRID 47640014


Table D.8. CBW mortality on artificially infested plots during the 2008 growing season
Event
                                  Total eggs/
                                 3 locations*
                                  Total fruit
                                   (average)
                                  Total live
                                    larvae*
                                 % live larvae
                              Est. live larvae**
                                  % Mortality
T304-40, Cry1Ab
                                     19800
                                     30422
                                       8
                                     0.84
                                    322.97
                                     99.52
GHB119, Cry2Ae
                                     19800
                                     24996
                                       3
                                     0.27
                                     86.40
                                     99.68
TwinLink(TM)
                                     19800
                                     32279
                                       7
                                     0.89
                                    309.01
                                     98.26
* 6,600 eggs were applied per plot; 1 plot = 1 location
Modified from MRID 47640014


   2. Pink Bollworm

A similar artificial infestation study as for CBW was carried out for PBW by Drs. Ellsworth and Li at the University of Arizona. Four plots were planted as a random complete block design (8 replicates) with TwinLink(TM), each single event, and Coker312 as control. Each plot was six rows long (30-ft) and separated from other blocks by one empty row. Young bolls were infested with 100 eggs each from a susceptible PBW strain maintained by USDA-ARS (Maricopa, AZ). Six to eight days later tagged bolls were brought into the lab for dissection. Attack rates on bolls were evaluated by counting the number of entry holes on the exterior of the boll carpel. PBW survival was estimated 6-15 days later which provided enough time for larvae to develop to 3[rd] instar. Survival was determined based on the number of exit holes, 3[rd], live 4[th], and pupae. Mortality (synonymously used with efficacy) was estimated by dividing the number of surviving PBW by the number of entry holes and by correcting for control mortality. Mortality in the Coker line was determined by limiting examinations to mainly those bolls with < 20 entry holes in order to avoid density-dependent mortality in control treatments. 

BCS concluded that all three infestations with Cry1Ab and TwinLink(TM) Cotton demonstrated high efficacy (corrected mortality >99.5%) against PBW. Two infestations with Cry2Ae demonstrated high efficacy (corrected mortality >99.1%), while results from one infestation were somewhat lower (corrected mortality 98.7%). BCS hypothesized that the small sample size from the first infestation in addition to the high control mortality may have contributed to this lower mortality estimate. Since control mortalities were high (Table D.9), corrected mortality rates based on each infestation showed somewhat lower levels of efficacy for each line. Cry2Ae was somewhat less effective in all infestations. When all data were pooled for each cotton line, the mortality rates were again very high (Table D.10).


Table D.9. Mortality results of susceptible PBW following artificial infestation of cotton plants
Infestation/ Data collection
                                   Treatment
                                  Entry Holes
                               Live Large Larvae
                             Average Mortality (%)
                                   Lower 95%
                                     Mean
                                Upper 95% Mean
                            Corrected Mortality (%)
July
                                     Coker
                                    Cry1Ab
                                    Cry2Ae
                                  TwinLink(TM)
                                     1646
                                     2756
                                     1901
                                      331
                                      726
                                       8
                                      11
                                       2
                                     50.26
                                     99.70
                                     99.23
                                     99.96
                                     46.37
                                     99.46
                                     98.65
                                     99.87
                                     54.16
                                     99.94
                                     99.82
                                      100
                                       -
                                     99.34
                                     98.69
                                     99.86
August
                                     Coker
                                    Cry1Ab
                                    Cry2Ae
                                  TwinLink(TM)
                                     1262
                                     2725
                                     2485
                                     1897
                                      393
                                       0
                                       5
                                       0
                                     59.75
                                      100
                                     99.84
                                      100
                                     54.39
                                      100
                                     99.70
                                      100
                                     65.10
                                      100
                                     99.98
                                      100
                                       -
                                      100
                                     99.35
                                      100
September
                                     Coker
                                    Cry1Ab
                                    Cry2Ae
                                  TwinLink(TM)
                                      598
                                     2399
                                     1551
                                     1452
                                      182
                                       3
                                       4
                                       0
                                     63.29
                                     99.75
                                     99.65
                                      100
                                     56.63
                                     99.32
                                     99.19
                                      100
                                     69.95
                                      100
                                      100
                                      100
                                       -
                                     99.59
                                     99.15
                                      100
 Modified from MRID 47640014

Table D.10. Pooled mortality results of susceptible PBW following artificial infestation of cotton plants
Treatment
                                 Total Entries
                                    N Bolls
                               Live Large Larvae
                              Apparent Mortality
                                Lower 95% mean
                                Upper 95% Mean
                              Corrected Mortality
                                      (%)
Coker
                                     3506
                                      259
                                     1301
                                     56.3
                                     53.4
                                     59.2
                                      - -
Cry1Ab
                                     7880
                                      345
                                      11
                                     99.8
                                     99.6
                                     99.9
                                     99.6
Cry2Ae
                                     5937
                                      252
                                      20
                                     99.6
                                     99.4
                                     99.8
                                     99.1
TwinLink(TM)
                                     6680
                                      302
                                       2
                                     99.9
                                     99.9
                                      100
                                     99.9
Modified from MRID 47640014


High-Dose Verification Method #3:

   1. Pink Bollworm

In 2007 and 2008, an artificial infestation study was carried out by Drs. Ellsworth and Li at the University of Arizona with a homozygous resistant strain of PBW. A Cry1Ac-resistant PBW strain (AZP-R) was obtained from Dr. Bruce Tabashnik at the University of Arizona. That strain was resistant to Cry1Ac with 53% survival in response to 10ug of Cry1Ac powder/g of diet and demonstrated cross-resistance to Cry1Aa and Cry1Ab. The artificial infestation experiment was run in the lab with field collected plant tissue of Cockerline (control), TwinLink(TM), Cry1Ab, and Cry2Ae plant tissue.
 
Young cotton bolls were harvested (15-19 days old) after tagging. Each boll was maintained in a water pick bioassay system. To measure the efficacy of the transgenic cotton, bolls were infested with an estimated 15 homozygous PBW neonates. The same procedures were followed for attack rate estimates, boll dissection, and PBW survival and mortality were the same as described for the PBW artificial infestation study using verification method #4.
 
A total of 348 cotton bolls were assayed. The mean entry holes ranged from 6.2-28.5. Attack rates were much higher in some cases. Maximum number of entry holes on a boll were 31 (Coker), 48 (Cry1Ab), 48 (Cry2Ae), and 54 (TwinLink(TM) Cotton). The mean percentage of PBW recovered ranged from 80.1 to 94.5%, which was similar to assay results with susceptible PBW. The criterion for `survival' at the time of boll dissection was a `living >3[rd] instar larva'. Mortality was calculated by dividing the number of surviving PBW by the number of entry holes as well as by correcting for control mortality. JMP IN Statistical Discovery Software, Version 5.1, SAS Institute 2005 was used for the data analyses.

In two out of three infestations, Cry2Ae demonstrated 100% efficacy; TwinLink(TM) Cotton demonstrated 100% efficacy in all three infestations. Cry1Ab treatments were less effective in the first and second infestation and demonstrated slightly better efficacy in the third infestation. However, corrected mortalities for Cry1Ab were similar and just below 50%. When all data were pooled for each treatment (or cotton line), the mortality rates were high. PBW mortalities in Cry2Ae and TwinLink(TM) were not significantly different from 100%. The reported cross-resistance of AZP-R to Cry1Ab toxin was consistent with the findings in this study. No reported cross-resistance to Cry2Ae was apparent. The authors concluded that Cry2Ae and TwinLink(TM) Cotton expressed a high-dose against PBW.

The authors stated that PBW survival on Bt plants (TwinLink(TM) and Cry2Ae Cotton) could have been higher because 1[st] instar larvae might have been exposed to hemizygous rather than purely homozygous seed tissue. Experimental Bt plot sizes where small, and Coker null lines were in close proximity, which could have facilitated low rates of out-crossing. Also, other factors like environment or altered production of protein due to warting (after bollworm attack) may have led to lower Bt expression in bolls than expected and hence, led to higher survival of PBW on TwinLink(TM) and Cry2Ae plants.

Table D.11. Mortality results for Cry1Ac resistant PBW following artificial infestation of cotton plants
Infestation/ Data collection
                                   Treatment
                                  Entry Holes
                               Live Large Larvae
                            Apparent Mortality (%)
                                   Lower 95%
                                     Mean
                                Upper 95% Mean
                            Corrected Mortality (%)
July
                                     Coker
                                    Cry1Ab
                                    Cry2Ae
                                  TwinLink(TM)
                                      210
                                      466
                                      448
                                      569
                                      86
                                      99
                                       1
                                       0
                                     51.9
                                     76.8
                                     99.8
                                      100
                                     36.0
                                     71.3
                                     99.4
                                      100
                                     67.7
                                     82.3
                                      100
                                      100
                                       -
                                     48.1
                                     99.5
                                      100
August
                                     Coker
                                    Cry1Ab
                                    Cry2Ae
                                  TwinLink(TM)
                                      279
                                      431
                                      467
                                      468
                                      89
                                      71
                                       0
                                       0
                                     67.1
                                     79.8
                                      100
                                      100
                                     60.5
                                     73.1
                                      100
                                      100
                                     73.7
                                     86.5
                                      100
                                      100
                                       -
                                     48.4
                                      100
                                      100
September
                                     Coker
                                    Cry1Ab
                                    Cry2Ae
                                  TwinLink(TM)
                                      111
                                      71
                                      141
                                      327
                                      30
                                       9
                                       0
                                       0
                                     67.5
                                     90.3
                                      100
                                      100
                                     52.6
                                     78.6
                                      100
                                      100
                                     82.4
                                      100
                                      100
                                      100
                                       -
                                     53.1
                                      100
                                      100
Modified from MRID 47640014


Table D.1. Pooled mortality results for Cry1Ac resistant PBW following artificial infestation of cotton plants
Treatment
                                 Total Entries
                                    N Bolls
                               Live Large Larvae
                              Apparent Mortality
                                Lower 95% mean
                                Upper 95% Mean
                              Corrected Mortality
                                      (%)
Coker
                                      600
                                      56
                                      205
                                     64.5
                                     58.3
                                     70.7
                                       -
Cry1Ab
                                      968
                                      74
                                      179
                                     80.3
                                     76.0
                                     85.6
                                     45.9
Cry2Ae
                                     1056
                                      85
                                       1
                                     99.95
                                     99.85
                                      100
                                     99.7
TwinLink(TM)
                                     1364
                                      112
                                       0
                                      100
                                      100
                                      100
                                      100
Modified from MRID 47640014



IV.       CROSS-RESISTANCE POTENTIAL (MRID 47640014)

BCS referred to a series of binding studies (Karim et al., 2000; Luo et al., 2007; Hernandez-Rodriguez et al., 2008) to support that Cry1 and Cry2 proteins bind to different target sites in the larval midgut of Lepidoptera pests. In addition, work by Hernandez-Rodriguez et al. (in press) showed that Cry2Aa, Cry2Ab, and Cry2Ae share a common binding site in CBW and the Old World bollworm. 

BCS commissioned Dr. Jurat-Fuentes at the University of Tennessee to conduct competitive binding experiments with radio labeled Cry1Ab and Cry2Ae proteins using BBMV of TBW and CBW to confirm that these proteins bind to the target pests' midgut with specificity. Furthermore, Dr. Jurat-Fuentes conducted a study with Cry1F, Cry1Ab, Cry1Ac, Cry2Ab, and Cry2Ae to fully characterize competitive binding aspects between the two newly proposed Bt toxins for cotton and those toxins already found in already registered Bt cotton products (see figures and results below).

BCS did not conduct binding studies for the third main target pest, PBW; rather the applicant cited publicly available literature. In experiments by Karim et al. (2000), Cry1Ab and Cry1Ac bound specifically and with high affinity to BBMV of PBW, while Cry1Aa and Cry2Aa bound with less specificity and lower affinity yet showed high binding site concentration. BCS stated that these results confirmed that Cry1A and Cry2A proteins were unlikely to be for cross-resistant, as was seen in other insect studies.

   1. Binding Experiments with Cry1Ab and Cry2Ae using CBW Brush Border Membrane Vesicles
   
The BBMV experimental results showed that labeled Cry1Ab (see Figure 2) and Cry2Ae (not shown here) bound with specificity to the midgut of CBW.

















Figure D.2. BBMV binding assays with increasing CBW  BBMV concentrations and fixed radio-labeled Cry1Ab (from MRID47640014). Unlabeled and labeled toxin (of constant amount) tend to bind to specific sites first; once the specific receptors are occupied, toxins bind to unspecific sites. Non-specific binding continues to increase linearly because the concentration of non-specific sites increases with increasing BBMV.
      

   2. Binding Experiments with Cry1Ab and Cry2Ae using TBW Brush Border Membrane Vesicles

The BBMV experimental results showed that labeled Cry1Ab (see Figure 3) bound with specificity to the midgut of TBW. The results for Cry2Ae are not shown here.

















Figure D.3. BBMV binding assay with increasing TBW BBMV concentrations and fixed radio-labeled Cry1Ab (from MRID47640014)

   
   3. Competition Experiments between Cry1Ab, Cry1Ac, Cry1F, Cry2Ab, and Cry2Ae using CBW Brush Border Membrane Vesicle

A competitive binding experiment with Cry1Ab, Cry1Ac, Cry1F, Cry2Ab2, and Cry2Ae showed that Cry1Ab shared high affinity binding sites with Cry1Ac and Cry1F in CBW. However, Cry1Ab did not compete with Cry2Ab and Cry2Ae suggesting that these proteins do not share binding sites in CBW (See Figure 4).
















Figure D.4. Competition binding of Cry1Ab and competitors to BBMV of CBW (from MRID 47640014); if two different Bt toxins share binding sites, there will be less binding by the labeled toxin (with constant concentration) as the amount of competitor increases. When there is no competition, a flat line across can be expected for the competitor concentration.


   4. Competition Experiments between Cry1Ab and Cry2Ae using TBW Brush Border Membrane Vesicles

Results of a binding competition experiment between Cry1Ab and Cry2Ae using TBW BBMV showed that the two toxins did not share binding sites (Figure 5).

















Figure D.5. Competition assay with Cry1Ab and Cry2Ae using BBMV of TBW

In conclusion, BCS stated that the relative potential for cross resistance was low between variants of Cry1 and Cry2 toxins as expressed in Bt cotton. Yet, the potential for cross-resistance was high within the group of Cry1 variants (e.g. Cry1Ab, Cry1Ac, and Cry1F) as well as within the group of Cry2 variants (e.g. Cry2Ab and Cry2Ae) (see Table below).


Table D.2. BCS' estimated potential for cross-resistance between various Bt proteins in TBW, CBW, and PBW
                                     Toxin
                                    Cry1Ab
                                    Cry1Ac
                                     Cry1F
                                    Cry2Ab
                                    Cry2Ae
Cry1Ab
                                      N/A
                                      +++
                                      ++
                                       -
                                       -
Cry1Ac
                                       
                                      N/A
                                       +
                                       -
                                       -
Cry1F
                                       
                                       
                                      N/A
                                       -
                                       -
Cry2Ab
                                       
                                       
                                       
                                      N/A
                                      +++
      Modified from MRID 47640014  
      +++, ++ = high potential; + = low potential; - = little to no potential


V.       MODELING 

BCS commissioned Dr. Mike Caprio (Mississippi State University) to conduct a risk assessment for TwinLink(TM) cotton on CBW using a model by Caprio et al. (2009). Dr. Caprio conducted a probabilistic risk assessment using a deterministic model and using PERT distributions (type of Beta distribution with mean was weighted, for example, 4x more than minimum and maximum values) for parameters of interest (with the exception of two cases: dominance associated with fitness costs). One thousand simulations were run with reasonable parameter values chosen at random from distributions derived with expert feedback (Table D.14).

The model included 43 different habitats from early season wild hosts, non-Bt corn, Bt cotton, soybeans, sorghum, and late season wild hosts. A field of cotton progressed through four different habitats in a growing season: fallow field (unsuitable habitat), pre-bloom cotton, flowering cotton, peak bloom cotton, and back to fallow field. Two Bt traits were modeled, Cry1Ab was assumed to have complete cross-resistance to Cry1Ac, and Cry2Ae was assumed to have complete cross-resistance to Cry2Ab2.

The size of each field was assumed to be 100 acres, and the modeling landscape consisted of 1024 fields. Corn, cotton, and soybeans accounted for 260 fields each, sorghum for 99 fields, and wild hosts for 145 fields. 100% of cotton was simulated as TwinLink(TM) cotton. No Cry1Ab corn was modeled in the landscape because BCS did not have access to sales information.

The rate of resistance evolution was estimated in two ways: (1) The final R-allele frequency after 31 years was divided by the initial frequency (was not constant due to re-sampling) and the result was log transformed. In many cases, the resistance allele frequency exceeded 90% and, once there, could no longer increase. All cases that approach the frequency of 1.0 received the same score no matter how many years it took to approach that value. (2) The rate of change in the resistance allele frequency was measured after three years to estimate the rate of resistance evolution. 


Table D.3. Parameters varied and used to define probability function, and function used for random draws
Parameter
                                      Max
                                      Min
                                  Most Likely
                                 Distribution
Initial R-allele frequency, Cry2Ae
                                     0.01
                                    0.00001
                                    0.0002
                                   Pert-Log
Initial R-allele frequency, Cry1Ab
                                     0.05
                                    0.00001
                                     0.002
                                   Pert-Log
R-allele fitness cost (independently drawn for both alleles)
                                      0.5
                                      0.0
                                      0.1
                                     Pert
R-allele fitness cost dominance (independently drawn for both alleles)
                                      0.2
                                      0.0
                                     0.05
                                  Mpert (2.0)
Mortality from Cry2Ae
                                    0.9999
                                     0.999
                                     0.99
                                   Pert-Log
Mortality from Cry1Ab
                                    0.99999
                                    0.9999
                                     0.99
                                   Pert-Log
Dominance of resistance (independently drawn for both alleles)
                                      0.3
                                     0.00
                                     0.05
                                     Pert
Modified from MRID 47640014

In the model, resistance did not evolve to either toxin in TwinLink(TM) cotton in 1000 simulations of 31 years of use; hence, under the conditions modeled there appeared little risk of product failure in a 30-year time horizon. In none of the simulations did the resistance allele frequency increase more than three-fold (see Figures 6 and 7); in most cases the R-allele frequency decreased. The fitness costs simulated were low (recessive in nature), however, the effects of fitness costs exceeded the effects of selection. The report was concluded with "These results support the contention that even small fitness costs can have a much larger impact on the evolution of resistance to stacked toxins than the case of plants expressing a single toxin".


Figure D.6. Rate of increase in Cry1Ab resistance allele frequency after three years. Decreases in the resistance allele frequency occur when fitness costs exceeded increases due to selection (taken out of MRID 47640014)


Figure D.7. Rate of increase in Cry2Ae resistance allele frequency after 3 years

VI.       REFUGE STRATEGY

BCS stated that since 1) both Cry1Ab and Cry2Ae express high-dose for the major target pests, CBW and TBW, and 2) the two toxins bind to different target sites in the midgut of the three insect pests, that TwinLink(TM) cotton should qualify for Natural Refuge. Their request was accompanied by a letter from Monsanto Company which gave BCS permission to cite relevant studies in support of a natural refuge request. 

A waiver request for an IRM plan was submitted to the Agency because BCS does not intend to commercially sell seeds from the single Events Cry1Ab and Cry2Ae. Very small acreages may be grown for seed production only in order to test combinations with other insect control traits that could potentially be candidate commercial products.

BCS submitted a structured refuge request for the target pest pink bollworm.

VII.       RESISTANCE MONITORING PROGRAM

BCS proposed to monitor for resistance evolution in all three target pests in areas of high TwinLink(TM) adoption.

The applicant proposed to have baseline susceptibility established for Cry1Ab and Cry2Ae proteins by 2012 (for CBW, TBW, and PBW) prior to commercialization. Sample collection for TBW and CBW would occur in the following manner: 10-20 populations for each pest would be collected in the primary cotton growing regions of the USA. For PBW, 4-6 populations would be collected from Arizona, New Mexico, West Texas, and California. This preliminary susceptibility data will provide valuable baseline information to compare with future annual insect susceptibility monitoring data. BCS will also use this baseline information to develop a discriminating dose for each toxin and each pest.

Once TwinLink(TM) has been commercialized, BCS has committed to monitoring of the three target pests on an annual basis.


VIII.       REMEDIAL ACTION PLAN

BCS stated that if insect resistance is suspected (monitoring or unexpected crop damage reports), it will attempt to confirm and characterize the resistance. First, the damaged cotton plants would be tested to confirm that the Bt proteins was produced and that a target pest of TwinLink(TM) cotton was responsible for the observed damage. If the plants were confirmed to express TwinLink(TM), the insect damage was from a primary target pest, and all other reasonable causes for the field damage had been ruled out (i.e. weather, cultural reasons), then insect populations would be collected from the region. Protein and plant bioassays would be conducted to determine the level and extent of resistance on the sampled insect populations and a colony established to determine if the putative resistance was heritable. Data from standardized bioassays would be compared to the original baseline data. Communication with customers and extension agents in the region, the EPA, and other registrants and applicants, whose products could be impacted by the findings, is another aspect of BCS' proposed remedial program.

Confirmed resistance would define the remedial action area(s) as well as the approach. The main focus would be to limit the spread of resistant alleles by eliminating resistant population(s) as quickly as possible. This might be accomplished through chemical insecticide treatments of the Bt cotton field effective against the resistant insects and/or using crop destruction methods that could limit the growth and reproduction of the resistant population(s). Susceptibility monitoring would be enhanced, especially in the region where resistance was identified. Refuge acreage may be modified and/or suspension of product sales could occur in the affected areas, not only TwinLink(TM), but for all products expressing similar Bt proteins that could be potentially cross-resistant.

IX.       GROWER EDUCATION

BCS stated that it will use grower meetings to introduce the TwinLink(TM) technology and guide the growers on best practices to capture the highest production yields. Growers have the responsibility to monitor for unexpected crop damage in TwinLink(TM) and to report product failure to BCS. A toll free telephone number will be provided in brochures and other communication tools for efficient tracking and follow-up of unexpected insect damage reports.

BCS will ensure that IRM plans are fully communicated to BCS sales staff and distribution channels and initially, will limit TwinLink(TM) seed sales to its own seed company, under FiberMax and Stoneville varieties. This will provide BCS with oversight on communication and control with educational efforts of sales staff. The transition to a natural refuge for combined trait Bt cotton products will reduce the education/communication needs related to refuge deployment and eliminate the need to monitor grower compliance with IRM requirements aside from the required field monitoring for unexpected insect damage. 

X.       GROWER COMPLIANCE

Grower compliance with refuge and IRM requirements is a critical element for resistance management. Significant non-compliance with IRM among growers may increase the risk of resistance for Bt crops. To minimize the effects of non-compliance, it is necessary to develop a broad compliance program as part of the IRM strategy. Such a program has to include 1) an understanding of the effect of non-compliance on IRM; 2) identification of compliance mechanisms to maximize adoption of IRM requirements; 3) measurement of the level of compliance; and 4) establishment of an enforcement structure to ensure compliance and penalize non-compliance.

XI.       REPORTING REQUIREMENTS 

Once the baseline susceptibility data are collected for the three primary target pests and a discriminating dose level has been established, annual monitoring will be conducted and reported to the agency. 

XII.       BPPD IRM REVIEW 

A. PROTEIN EXPRESSION
   
BPPD reviewed BCS' protein expression results reported in their IRM submission and concludes that single Event T304-40 may not express the same levels of toxin as in the dual gene PIP, TwinLink(TM). It appears that T304-40 was expressed at higher levels in leaves of single event plants compared to leaves collected from TwinLink(TM) plants. For flowers and bolls the pattern was reversed, and TwinLink(TM) plants appeared to express T304-40 at higher levels in these tissues. It is unclear whether these differences in expression results were statistically significant because BCS did not provide further information on the differences observed. 

Expression of GHB119 in leaves from early to late stage appeared somewhat variable or unstable. Expression levels were measured in leaves in 2004 and 2005 at three developmental stages (early leaf, mid leaf, and late leaf). At developmental stage 1, expression was higher than at developmental stage 2, and at developmental stage 3, expression increased again but to higher levels than observed at developmental stage 1. For early, mid, and late boll, Expression level for GHB119 were also assessed in three different developmental stages of bolls (early, mid, and late). Levels of GHB119 were higher in TwinLink(TM) than in the single event plant during developmental stage 2 and 3. It appears that the difference was approximately two-fold (see Figure D.1). 

Table D.1 in this review summarizes the protein expression level (ug/g-dry weight) submitted by BCS for T304-40 and GHB119 in different plant tissues. It appears that the same tissue types were lumped together for plants expressing single and pyramided events. Since there appears to be unequal expression of the proteins in plants expressing either the single or pyramided events over the course of the growing season, this approach is not providing the most information. It would be more useful to have the expression data broken down by tissue for each individual Event, GHB119, T304-40, and TwinLink(TM), rather than lumped together for the same protein.

BPPD recommends that BCS provide more specific and explanatory information together with their protein expression results in future data submissions for PIP registrations.
 


B. DOSE

The determination of dose, or the amount of toxin expressed by the transgenic crop relative to the susceptibility of the target pests, is a critical component of IRM. Models have shown that a high-dose of toxin, coupled with a non-transgenic refuge to provide a supply of susceptible insects, is the most effective strategy for delaying resistance in Bt crops. The high-dose/refuge strategy assumes that resistance to Bt is recessive and is conferred by a single locus with two alleles resulting in three genotypes: susceptible homozygotes (SS), heterozygotes (RS), and resistant homozygotes (RR). The high-dose/refuge strategy also assumes that there will be a low initial resistance allele frequency and extensive random mating between resistant and susceptible adults. In practice, a high-dose PIP should express sufficient quantities of toxin to kill all susceptible insects (SS) as well as heterozygous insects with one resistance allele (RS). Lower dose PIPs might allow for survival of insects with at least one susceptibility allele (SS or RS), although effective IRM may still be possible with a suitable refuge strategy. To be able to demonstrate high-dose, it is recommended that applicants generate data by at least two of the five laboratory and field approaches as outlined by the SAP (1998) and described by the Agency in the 1998 Bt Plant-Pesticides and Resistance Management document (US EPA, 1998) and 2001 Biopesticide Registration Action document (US EPA, 2001).

The high-dose definition and verification techniques were developed in 1998 when all of the registered Bt crops were single toxin products targeted against Lepidopteran pests. In recent years, PIPs in Bt cotton have been approved that contain two genes targeted at the same insect pest. These "pyramided" products can be beneficial for IRM since target pests must overcome two toxins to develop field resistance to the PIP. The benefits are greatest for two toxins with unrelated modes of action (i.e. binding to different Bt receptor sites in the midgut) that are expressed at high-doses in the plant (Roush 1998). 

For pyramided products, the dose of each toxin should be evaluated separately. In cases where the pyramided product is created through conventional breeding and the dose of the single toxin products has already been established, the combined dose in the pyramided PIP can be determined with comparative efficacy studies and protein expression data. However, this is not the case for TwinLink(TM) cotton.

BPPD reviewed BCS' dose information for TBW, CBW, and PBW and concludes the following:
   
   1. Tobacco Bollworm

BCS submitted one dose verification method for TBW rather than two. Since all verification methods may be imperfect at assessing dose against a target pest when used alone, the Scientific Advisory Panel (1998) recommended that at least two methods (of five methods described by the panel) be used when demonstrating dose expression in target pests. BPPD notes that a second dose verification method is lacking for TBW. In the past Natural Refuge requests were granted on the basis that the toxins expressed high-dose against TBW. A second dose-verification method will need to be submitted if BCS wishes to maintain that GHB119, T304-40, and TwinLink(TM) express a high-dose against TBW and to support their Natural Refuge request. 

High-dose verification method (1):
From mortality results obtained with dose verification method (1), BPPD concludes that Events GHB119, T304-40, and TwinLink(TM) may express a high-dose and effective high-dose, respectively, against TBW. 

After day 4 of the diet-incorporated bioassay, mortality was reported as 100% for TwinLink(TM) and Cry2Ae treatments, while control mortality ranged from 0%-2%. For the overlay assay, mortality was 100% for TwinLink(TM) and Cry2Ae by day 7, while control mortality was 15%.

For Cry1Ab treatments, mortality reported from diet bioassays did not reach 100% until after day 15. In addition, control mortality was relatively high (ranging from 23%-28%). For overlay assays, mortality for TBW when exposed to Cry1Ab was 100% by day 15; control mortality at that time had reached 17%. Cry1Ab may express a high-dose against TBW. 

   2. Cotton Bollworm 

It appears that TwinLink(TM) may express a possible effective high-dose against CBW. However, a second dose verification method turned out to be inconclusive during BPPD's review and therefore, a high-dose cannot be verified. Should BCS wish to maintain that TwinLink(TM) expresses a high-dose against CBW, additional information needs to be provided to clarify uncertainties in the experiment using high-dose verification method 4. 
   
High-dose verification method (1): 
BPPD concludes that the diet-incorporated and overlay assays together suggest that TwinLink(TM) and GHB119 may express a high-dose, but T304-40 may express less than high-dose against CBW.

In the control treatments, larvae had developed into pupae by day 17. BCS did not report whether surviving larvae of the exposure treatments, Cry1Ab, Cry2Ae, and TL, developed into pupae and consequently adults or whether survivors remained in the larval stage. BPPD notes that for future submission, it is recommended that BCS report whether surviving larvae were developmentally `dead' (as opposed to physiologically), or whether they continued to mature into pupae and adults.

High-dose verification method (4): 
BPPD identified several uncertainties in the methodology section that preclude making a dose determination based on study results obtained with dose verification method (4).

BCS stated in the IRM submission that mortality results were obtained from an artificial and natural infestation experiment done in conjunction with the efficacy study described in chapter 13 of their submission (MRID 47640012). When BPPD consulted chapter 13 for raw data or data tables, however, mortality data from the infestation experiments were lumped for all lepidopteran pests, and no pest specific mortality data were reported. It is unclear how the data that BCS reported in chapter 15 of their submission (Tables 7 and 8) were obtained, and no further details about the study were submitted in an appendix accompanying their submission (MRID 47640014).

Furthermore, BCS stated in study volume 13 that some plots were treated with conventional pesticides, while others were not. It is unclear whether the mortality results reported in Table 7 and Table 8 come from plots that have been sprayed, not sprayed, or both, and what effect spraying might have had on dose determinations. 

Coker cotton was planted in the control plot treatments for this natural and artificial infestation experiment; however, no information on control mortality was provided in the submitted report. It cannot be determined whether the final CBW mortality results from plots expressing Cry1Ab, Cry2Ae, and TwinLink(TM) have been corrected for control mortality. 

The applicant may have conducted an egg hatchability study along with the infestation experiment to estimate egg viability and larval attack rate. BPPD has not been provided with egg hatchability data or other information that would allow estimating viability and attack rate. Without such information it is unclear how many larvae hatched, attacked the Bt cotton plants, and ultimately survived Bt exposure. The applicant did, however, provide the infestation rate per plant.

Furthermore, BCS provided BPPD with CBW average mortality across two locations in 2007, and three locations in 2008, but no results were provided for individual locations. In addition to providing average mortality results, BPPD requests that individual treatment mortalities be reported for future data submissions so that minimum mortality per toxin is known. 

Should BCS wish to pursue the high-dose claim for TwinLink(TM) and CBW, BPPD recommends providing clarification and further data for the study using high-dose verification method (4).

BPPD further recommends that the applicant provide more in-depth descriptions of results and methodology for future studies and data submissions.

   3. Pink Bollworm 

High-dose verification methods (3) and (4):
BPPD cannot make a conclusion about the dose profile for TwinLink(TM) and its events and PBW (using verification methods 3 and 4) because it is unclear how the mortality results reported by Ellsworth and Li (2008) were obtained. When BPPD divided the number of entry holes reported (Tables 9-12) by the number of live larvae and calculated average mortality, the obtained results did not match mortalities reported by BCS. 

Deficiency 1: The dose profile reported for PBW cannot be verified by BPPD until BCS clarifies how mortalities were calculated for the experiments using dose verification methods (3) and (4).


Table D.4. BPPD Conclusion Regarding Dose Profile for Events T304-40, GHB119, and TwinLink(TM)
                                     Pest
                                   Method #1
                                   Method #4
                                   Method #3

                                    T304-40
                                    GHB119
                                      TL
                                    T304-40
                                    GHB119
                                      TL
                                    T304-40
                                    GHB119
                                      TL
CBW
                                 Non-High Dose
                              Possible High-Dose
                         Possible Effective High-Dose
                                   Inconcl.*
                                   Inconcl.*
                                   Inconcl.*
                                      --
                                      --
                                      --
TBW
                              Possible High-dose
                                   High-Dose
                              Effective High-Dose
                                      --
                                      --
                                      --
                                      --
                                      --
                                      --
PBW
                                      --
                                      --
                                      --
                                   Inconcl.*
                                   Inconcl.*
                                   Inconcl.*
                                      ---
                                   Inconcl.*
                                   Inconcl.*
* Insufficient information was submitted to the Agency to make a dose determination 
** Cry1Ab could not be tested in this experiment because the lab strain was resistant to Cry1Ab. 
   
C. CROSS-RESISTANCE

Analyses of resistance to Bt Cry proteins indicate that cross-resistance occurs most often with proteins that are similar in structure (Tabashnik, 1994; Gould et al., 1995). BCS cited the work by Crickmore et al. (1998), which supports that Cry1 and Cry2 Bt proteins do not have more than 20% amino acid sequence homology in common. Furthermore, the 3-dimensional crystal structures of Cry2Aa (very closely related to Cry2Ae as expressed in TwinLink(TM)) and Cry1Aa (very closely related to Cry1Ab as expressed in TwinLink(TM) proteins have been resolved by Morse et al. (2001) and demonstrate that the proteins have a distinct tertiary structure. This significant difference in structure suggests that these two classes of proteins may also have unique biochemical mechanisms, specifically with respect to binding sites. The similarity in the global architecture of Cry2Aa and Cry1Aa, however, suggests that the two toxins may have similar general steps for the mode of action.

BPPD concludes that the evidence provided by BCS supports that cross-resistance between Cry1Ab and Cry2Ae, as expressed in TwinLink(TM) cotton, is unlikely to develop in the main target pests CBW, TBW, and PBW. To summarize, the amino acid sequence homology between the two Bt toxins is less than 20%; the two protein structures differ substantially at the third terminal suggesting that the two proteins may have unique mechanisms with respect to binding sites (and other biochemical pathways); the binding assay studies by Jurat-Fuentes showed that Cry1Ab and Cry2Ae were not competing for the same binding sites in CBW as well as TBW; and experiments by Karim et al. (2000) demonstrated that Cry1Ab binding to gut membrane of PBW was specific while Cry2A binding was less specific yet with high binding site concentrations (suggesting that toxins were not competing for same binding sites).

However, BPPD notes that should resistance evolve in CBW and TBW to Cry2Ae as expressed in TwinLink(TM) cotton that Cry2Ab2 (as in Bollgard II) may be compromised as well because, as BCS stated in their report, there is a great potential for cross-resistance between the  two toxins. 

BPPD notes that the cross-resistance potential is great between Cry1Ab, Cry1Ac, and Cry1F (although somewhat less for Cry1F) based on the competition assays; the three toxins share high affinity binding sites. Should CBW evolve resistance to Cry1Ab, other PIP products expressing Cry1Ac and Cry1F will likely be compromised as well. 

D. MODELING

EPA has used predictive models to compare IRM strategies for Bt crops. Because models cannot be validated without actual field resistance, models have limitations and the information gained from the use of models is only a part of the weight of evidence used by EPA in assessing the risks of resistance development. It was the consensus of the SAP (2000) Subpanel that models were an important tool in determining appropriate Bt crop IRM strategies. They agreed that models were "the only scientifically rigorous way to integrate all of the biological information available, and that without these models, the Agency would have little scientific basis for choosing among alternative resistance management options." They also recommended that models must have an agreed upon time frame for resistance protection. For example, conventional growers may desire a maximum planning horizon of five years, while organic growers may desire an indefinite planning horizon. The Subpanel recommended that model design should be peer reviewed and parameters validated. Models should also include such factors as level of Bt crop adoption, level of compliance, economics, fitness costs of resistance, alternate hosts, spatial components, stochasticity, and pest population dynamics.

BPPD reviewed the modeling assessment and accepted the probabilistic approach presented therein. Based on the choice for parameter values, natural refuge assumptions, and their analysis, BCS stated that there is little risk of resistance evolution in CBW to TwinLink(TM) within 30 years. BPPD needs further clarifications on several issues and has some concerns regarding assumptions made for certain important parameters. These are summarized below.

Deficiency 2: In the modeling analysis provided by BCS, only CBW was modeled but not TBW. For a new refuge strategy, such as Natural Refuge, each main target pest needs to be evaluated separately via a simulation model that incorporates biology, genetics, toxicity, and ecology of the target pest. BCS needs to provide simulation modeling for TBW in order to support their natural refuge request.

Deficiency 3: The natural refuge paradigm was approved for Bollgard II and WideStrike with a high-dose profile for TBW. If BCS wishes to pursue the natural refuge strategy for TwinLink(TM), then a second high-dose verification study will need to be submitted to support that the PIP expresses a high-dose against TBW, or BCS can design a model with their specific dose data for TBW.

Deficiency 4: CBW is the pest of most concern since its populations show greater genetic variability to toxins expressed in Bt cotton (Cry1Ac, Cry1F, and Cry2Ab2) than other pest (i.e. TBW). BCS's LC50 data show that CBW have a wide range of susceptibility against Cry1Ab (462 -2844 ng/cm[2]) as expressed in cotton when compared to TBW (4.9-10.5 ng/cm[2]). BCS concluded in their report that TwinLink(TM) expressed less than high-dose against CBW due to the target pest being less susceptible to Cry1Ab and lower level of expression of Cry1Ab in the parental event for TwinLink(TM). In addition, bioassay results for Cry1Ab in corn (BPPD 2009) showed greater variability as well for Corn earworm, CEW, (aka CBW) than for other pests, such as ECB. In fact, there is up to a three-fold difference in magnitude of reported LC50s of CEW compared to ECB. And previous attempts to create a diagnostic concentration for CEW were frustrated by the fact that the insect was less susceptible to Cry1Ab than other Lepidopteran pests. Given this information, BPPD is concerned with the dose assumptions for CBW and recommends that an additional analysis be conducted using somewhat lower mortality assumptions based on the study results reported by BCS. The bioassay experiments showed that Cry1Ab caused mortality ranging 88.5-100% (see this review). Estimated mortalities from the artificial infestation experiments (2007-2008) ranged from 99.5-99.7% (see this review). However, BPPD cannot verify the results of the artificial infestation experiments due to lack of information provided (i.e. % egg viability). At the moment, there is no evidence that Cry1Ab expresses high-dose against CBW; hence BPPD recommends exploring dose mortalities ranging from 90-99% in the simulation model.

Deficiency 5: BPPD is concerned about the most likely value chosen for initial resistance allele frequency for Cry1Ab (0.0002). Although no field resistance to Cry1Ab has been documented yet for corn earworm (CEW), the toxin has been in use for over a decade (in Bt corn) and selection may likely have increased the frequency of potential resistance alleles. However, because corn earworm does not overwinter in the Corn Belt, BPPD agrees that the most likely value will be somewhat lower than could otherwise be expected. BPPD recommends that BCS consider modeling a most likely resistance allele frequency of 0.001 for Cry1Ab in CBW.

Deficiency 6: BCS needs to include Cry1Ab corn and Cry1Ab cotton in the modeling landscape when assessing the risk of resistance evolution in the target pests because Syngenta has registered Bt cotton and Bt corn and Monsanto has registered Bt corn products expressing Cry1Ab. These PIPs are represented in the Cotton Belt and need to be included in a simulation analysis for TwinLink(TM).

E. REFUGE STRATEGY

An IRM plan waiver request for single trait Events (Cry1Ab and Cry2Ae) is acceptable if seeds from those events are not sold commercially and serve only for producing the combination PIP TwinLink(TM). For any other combinations, BCS may need to apply for an experimental use permit or section 3 registration. BPPD recommends the use of acreage/geographic limitations for the single trait events as a means to mitigate potential resistance development. BCS should provide information on proposed acreage for these single trait events.

BPPD notes that a second dose verification method is lacking for TBW. In the past Natural Refuge requests were granted on the basis that the toxins expressed high-dose against TBW. A second dose-verification method will need to be submitted to support BCS' Natural Refuge request, or BCS can design a model with their specific dose data for TBW.

BPPD verified that BCS cited all the relevant natural refuge studies originally developed by Monsanto Company and has obtained a data citation permission letter from Monsanto.

When BCS has satisfactorily addressed the deficiencies related to the modeling of resistance evolution in CBW and TBW, BPPD will re-assess the natural refuge request for TwinLink(TM) cotton. 

If a Natural Refuge strategy for TwinLink(TM) is approved by BPPD for TwinLink(TM), the requirements for a non-Bt cotton refuge in conjunction with the planting of any acreage of TwinLink(TM) cotton should remain in place for the following states and regions: Arizona, California, and New Mexico and in the following Texas counties:  Brewster, Crane, Crockett, Culberson, El Paso, Hudspeth, Jeff Davis, Loving, Pecos, Presidio, Reeves, Terrell, Val Verde, Ward, and Winkler. 

For those States and regions requiring a non-Bt refuge, the following refuge planting requirements are recommended:

1)  5% External, Unsprayed Refuge

At least 5 acres of non-Bt cotton (refuge cotton) must be planted for every 95 acres of Bt cotton. The size of the refuge must be at least 150 feet wide, but preferably 300 feet wide. This refuge may not be treated with any insecticide labeled for the control of tobacco budworm, cotton bollworm, or pink bollworm. At the pre-squaring cotton stage only, the refuge may be treated with any lepidopteran insecticide to control foliage feeding caterpillars. A non-Bt cotton refuge has to be maintained within at least (1/2)  linear mile (preferably adjacent to or within 1/4  mile or closer) from the Bt cotton fields. 

2)  20% External Sprayed Refuge
At least 20 acres of non-Bt cotton are planted as a refuge for every 80 acres of Bt cotton (total of 100A). The non-Bt cotton may be treated with insecticides (excluding foliar Btk products) for control of the tobacco budworm, cotton bollworm, or pink bollworm. A non-Bt cotton refuge has to be maintained within at least 1 linear mile (preferably within (1/2) mile or closer) from the Bt cotton fields.
      
3)  5% Embedded Refuge 

The refuge cotton must be planted as at least one single non-Bt cotton row for every six to ten rows of Bt cotton. The refuge may be treated with any insecticide (excluding foliar Btk products) for the control of pink bollworm whenever the entire field is treated. The in-field refuge rows may not be treated independently of the surrounding Bt cotton field in which it is embedded. 

F. RESISTANCE MONITORING

The need for proactive resistance detection and monitoring is critical to the survival of Bt technology. The Agency mandates that applicants monitor for insect resistance (measurement of resistance-conferring alleles) to the Bt toxins as an important early warning sign to developing resistance in the field and whether IRM strategies are working. Grower participation (e.g., reports of unexpected damage) is also important for monitoring. Resistance monitoring is also important because it provides validation of biological parameters used in models. However, resistance detection/monitoring is a difficult and imprecise task. It requires both high sensitivity and accuracy. Good resistance monitoring should have well-established baseline susceptibility data prior to introduction of Bt crops. The chances of finding a resistant larva in a Bt crop depend on the level of pest pressure, the frequency of resistant individuals, the location and number of samples that are collected, and the sensitivity of the detection technique. Therefore, as the frequency of resistant individuals or the number of collected samples increases, the likelihood of locating a resistant individual increases (Roush & Miller 1986). If the phenotypic frequency of resistance is one in 1,000, then more than 3,000 individuals must be sampled to have a 95% probability of one resistant individual (Roush & Miller 1986). Current sampling strategies have a target of 100 to 200 individuals per location. Previous experience with conventional insecticides has shown than once resistant phenotypes are detected at a frequency >10%, control or crop failures are common (Roush & Miller 1986). Because of sampling limitations and monitoring technique sensitivity, resistance could develop to Bt toxins prior to it being easily detected in the field. (http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad.htm)

BCS' approach for developing baseline data for TBW, CBW, and PBW by 2012 may be too protracted in the event that TwinLink(TM) is commercialized before 2012. Baseline susceptibility data should be available by the first growing season of commercial use of the PIP product. BPPD recommends that BCS accelerate the development of these data. 

The applicant proposed to monitor for resistance evolution in all three target pests in areas of high TwinLink(TM) adoption. BPPD agrees with this proposal but also recommends that monitoring efforts additionally focus on regions in states with little natural refuge.

Deficiency 7: BCS needs to have a well-developed monitoring plan in place before TwinLink(TM) is approved for commercialization (i.e. sampling regions, sampling scheme, etc). BPPD recommends that BCS follow the monitoring paradigm established for the other Bt cotton registrations.

G. REMEDIAL ACTION PLAN

Remedial action plans are a potential response measure should resistance develop to Bt crops. 
Since resistance may develop in "localized" pest populations, it may be possible to contain the resistance outbreak before it becomes widespread. A specific remedial action plan should clearly indicate what actions the applicant will take in cases of "suspected" resistance (i.e., unexpected damage) and "confirmed" resistance. The remedial action plan can also include appropriate adaptations for regional variation and the inclusion of appropriate stakeholders. To fully mitigate resistance, a critical element of any remedial action plan should be that once pest resistance is confirmed, sales of all Bt cotton hybrids that express a similar protein or a protein in which cross-resistance potential has been demonstrated would be ceased in the affected region (see http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad.htm)

Deficiency 8: BCS needs to have an established remedial action plan in place for each target pest of TwinLink(TM). It is not sufficient to provide an outline or general statement about different kinds of approaches. BPPD recommends that BCS follow the existing remedial action paradigm for Bt cotton PIPs. 

BPPD notes that it will not be sufficient if BCS only attempts to confirm and characterize resistance upon receiving notification of unexpected damage. Rather, unexpected pest damage reports should be taken seriously and followed up immediately to eventually conclusively determine whether the cause of crop damage is due to resistance or other reasons. BPPD understands that it can take several months before a determination can be reached.

H. GROWER EDUCATION

It is recommended that BCS develop a grower education program that targets states and counties where a non-Bt corn variety must be planted as the refuge for TwinLink(TM) cotton. Those regions are Arizona, California, and New Mexico and the following Texas counties:  Brewster, Crane, Crockett, Culberson, El Paso, Hudspeth, Jeff Davis, Loving, Pecos, Presidio, Reeves, Terrell, Val Verde, Ward, and Winkler.

It is recommended that BCS develop a grower guide that at a minimum describes the technology and includes unexpected damage guidelines.

I. GROWER COMPLIANCE

BPPD notes that BCS should have a well developed grower-education program (including grower guide) for all regions with clear indications for regional requirements.

A natural refuge strategy for TBW and CBW would eliminate the need for growers to plant a refuge where alternate hosts are abundant. However, there are cotton-growing regions where natural refuge is not applicable and where planting a refuge is still required. For those regions it is recommended that BCS develop, implement, and report to EPA on programs to evaluate and promote growers' compliance with IRM requirements. These states and regions are Arizona, California, and New Mexico and in the following Texas counties:  Brewster, Crane, Crockett, Culberson, El Paso, Hudspeth, Jeff Davis, Loving, Pecos, Presidio, Reeves, Terrell, Val Verde, Ward, and Winkler.

Grower compliance in these regions should be assessed with on-farm assessments in addition to anonymous phone surveys. BPPD recommends that BCS use the existing grower compliance paradigm for other Bt cotton registrations.

J. REPORTING REQUIREMENTS

BCS stated that "once the baseline susceptibility data were collected for the three primary target pests and a discriminating dose level was established, annual monitoring would be conducted and reported to the agency. BPPD recommends that BCS submit annual reports for the following:

   * Baseline and discriminating dose work irrespective of their status;
   * Sales data;
   * Annual resistance monitoring; and 
   * CAP results for cotton growing regions where Natural Refuge is not applicable.
IRM REVIEW REFERENCES
     
Bushey, D. 2008. Request for a waiver from the requirement to develop an insect resistance management plan. Unpublished report submitted by BCS to the Agency, MRID 47634823

Caprio, M.A., Parker, C.D. and J.C. Schneider. Est. 2009. Future fitness of female insect pests in
temporally stable and unstable habitats and its impact on habitat utility as refuges for insect
resistance management. J. Insect Sci. in press

Crickmore, N., Zeigler, D.R., Feitelson, J., Schnepf, E, Van Rie, D., Lereclus, D., Baum, J., and Dean, D.H. 1998. Revisions of the Nomenclature for the Bacillus thuringiensis pesticidal crystal proteins. Microbiology and Molecular Biology Reviews, Vol. 62 (3): 807-813.

Gould, F., Anderson, A., Reynolds, A., Bumgarner, L., and Moar, W. 1995. Selection and genetic analysis of a Heliothis virescens (Lepidoptera: Noctuidae) strain with high levels of resistance to Bacillus thuringiensis toxins. J. Econ. Entomol. 88:1545-1559.

Hernández-Rodriguez, C.S., Van Vliet, A., Bautsoens, N., Van Rie, J. and J. Ferré (est. 2008) Specific binding of Bacillus thuringiensis Cry2A insecticidal proteins to a common site in the midgut of Helicoverpa sp. Appl. Environ. Microbiol. In press.

Karim, S., Riazuddin, S., Gould, F., and D.H. Dean. 2000. Determination of receptor binding properties of Bacillus thuringiensis δ-endotoxins to cotton bollworm (Heliocoverpa zea) and pink bollworm (Pectinophora gossypiella) midgut brush border membrane vesicles. Pest. Biochem. Physiol. 67:198-216.

Luo S., Wu K., Tian Y., Liang G., Feng X., Zhang J., and Y. Guo. 2007. Cross-resistance studies of Cry1Ac-resistant strains of Helicoverpa armigera (Lepidoptera: Noctuidae) to Cry2Ab. J. Econ. Entomol. 100:909-915.

MacIntosh, S., Holloway, J., Jansens, S., and Van Rie, J. 2008. Insect resistance management plan: TwinLink(TM) Cotton. Unpublished report submitted by Bayer CropScience to the Agency, MRID 47640014.

Roush, R. T., and G. L. Miller. 1986. Considerations for design of insecticide resistance monitoring programs. Journal of Economic Entomology, Vol. 79: 293-298.

Roush, R. 1996. Two-toxin strategies for management of insecticidal transgenic crops: can pyramiding succeed where pesticides mixtures have not? Phil. Trans. Royal Soc. Lond. B (1998): 1777- 1786

Tabashnik, B.E. 1994. Evolution of resistance to Bacillus thuringiensis. Annu. Rev. Entomol. 39:47-79. 







U.S. EPA, 1998. FIFRA Scientific Advisory Panel Subpanel on Bacillus thuringiensis (Bt) Plant-Pesticides and Resistance Management, February 9 and 10, 1998

U.S. EPA, 2000. FIFRA Scientific Advisory Panel Subpanel on Bt Plant-Pesticides Risk and Benefit Assessments: Insect Resistance Management, October 18-20, 2000

U.S. EPA, 2001. Biopesticides Registration Action Document  -  Bacillus thuringiensis Plant Incorporated Protectants, http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad.htm

BPPD. 2009. Review of 2007 monitoring data submitted by ABSTC for lepidopteran Bt corn registrants (MRID 47413901). Memorandum from S. Borges to M. Mendelsohn dated April 8[th], 2009.

Appendix D. II - EPA's Review of BCS' Response to Deficiencies 

Below is EPA's review of BCS' response to the eight IRM deficiencies noted in BPPD's risk assessment for TwinLink(TM) dated 4/22/2010 (EPA Reg. Nos. 264-RNOU, 264-RNOL, and 264-RNOA, MRIDs 4760014 and 47634823). On July 29[th], 2010, BCS submitted their response to BPPD's eight listed IRM deficiencies for TwinLink(TM) (BPPD 2010). BPPD reviewed BCS' response and additional information and concludes that five of the eight deficiencies were satisfactorily addressed and resolved (U.S. EPA 09/28/10). In BCS' response to deficiency 8, BPPD recommends including the following:  "reporting resistance to Agency within 30 days" and "a long-term IRM action plan" as outlined below. The applicant stated that the remaining deficiencies (#2 and #3) would be addressed during the next cotton growing season. BPPD recommends that these two items be included as terms of registration.

A) Discussion regarding Deficiency 1:

Deficiency 1: The dose profile reported for Pink bollworm (PBW) cannot be verified by BPPD until BCS clarifies how mortalities were calculated for the experiments using high-dose verification methods (3) and (4).

BCS' response: In high-dose verification experiments (method 3 and 4), BCS reported that the mortality was calculated as follows: Mortality = 1- number of large larvae/ number of entry holes. 

BPPD response: BPPD has verified the calculations and confirmed the reported mortality results. This deficiency has been satisfactorily addressed and resolved. BPPD concludes that the single Events Cry1Ab and Cry2Ae in cotton both express a less than high-dose against PBW and that TwinLink(TM) expresses a high-dose against PBW.


B) Discussion regarding Deficiencies 2-3:

Deficiency 2: In the modeling analysis provided by BCS, only cotton bollworm (CBW) was modeled but not Tobacco budworm (TBW). For a new refuge strategy, such as Natural Refuge, each main target pest needs to be evaluated separately via a simulation model that incorporates biology, genetics, toxicity, and ecology of the target pest. BCS needs to provide simulation modeling for TBW in order to support their natural refuge request.

Deficiency 3: The natural refuge paradigm was approved for Bollgard II and WideStrike with a high-dose profile for TBW. If BCS wishes to pursue the Natural Refuge strategy for TwinLink(TM), then a second high-dose verification study will need to be submitted to support that the PIP expresses a high-dose against TBW, or BCS can design a model with their specific dose data for TBW.

BCS' response: BCS informed BPPD that an additional high-dose study for TBW is scheduled for the next cotton growing season when field infestations are present. Modeling resistance evolution in TBW to TwinLink(TM) cotton will be conducted afterwards incorporating dose profile and Natural Refuge assumptions.

BPPD response: BPPD will review the dose data and modeling for TBW in support of a Natural Refuge request when the new data become available. In the meantime, in order to pursue registration with a structured refuge for TwinLink(TM), as proposed by BCS during the June 21 teleconference, this additional information is not required.

C) Discussion regarding Deficiencies 4-6:

Deficiency 4: BPPD is concerned with the dose assumptions for CBW and recommends that an additional analysis be conducted using somewhat lower mortality assumptions based on the study results reported by BCS. The bioassay experiments showed that Cry1Ab caused mortality ranging 88.5-100% (see this review). Estimated mortalities from the artificial infestation experiments (2007-2008) ranged from 99.5-99.7% (see this review). BPPD cannot verify the results of the artificial infestation experiments, however, due to lack of information provided (i.e. % egg viability). At the moment, there is no evidence that Cry1Ab expresses high-dose against CBW. BCS also concluded that TwinLink(TM) expressed less than high-dose against CBW due to the target pest being less susceptible to Cry1Ab and lower level of expression of Cry1Ab in the parental event for TwinLink(TM) Cotton. BPPD recommends exploring dose mortalities ranging from 90-99% in the simulation model.

Deficiency 5: BPPD is concerned about the most likely value chosen for initial resistance allele frequency for Cry1Ab (0.0002). Although no field resistance to Cry1Ab has been documented yet for corn earworm (aka CBW), the toxin has been in use for over a decade (in Bt corn) and selection may likely have increased the frequency of potential resistance alleles. However, because corn earworm does not overwinter in the Corn Belt, BPPD agrees that the most likely value will be somewhat lower than could otherwise be expected. BPPD recommends that BCS consider modeling a most likely resistance allele frequency of 0.001 for Cry1Ab in CBW.

Deficiency 6: BCS needs to include Cry1Ab corn and Cry1Ab cotton in the modeling landscape when assessing the risk of resistance evolution in the target pests because Syngenta has registered Bt cotton and Bt corn and Monsanto has registered Bt corn products expressing Cry1Ab. These corn and cotton PIPs are grown in the Cotton Belt and need to be included in a simulation analysis for TwinLink(TM).

BCS' response: The most likely R-allele value for Cry1Ab was actually set at 0.002 in the original submission; however, the initial allele frequency was set at 0.00001. In order to address BPPD's concerns, BCS increased the initial R-allele frequency of Cry1Ab to 0.0001. With respect to incorporating Cry1Ab corn in the modeling landscape, all corn crops were converted to 80% Bt (dual gene) corn and 20% non-Bt corn. Mortality rates were based on H. zea data on TwinLink(TM). Results after 2000 simulations (with revised parameters) showed that resistance did not evolve to either toxin in any of the simulations. The rate of increase over the first five years for each locus reflected the presence of fitness costs. In many cases, particularly with the Cry1Ab resistance locus, the resistance allele frequency actually dropped. This was likely due to the recessive nature of the fitness costs simulated. The most likely initial resistance allele frequency was 10-fold greater for this locus than for the Cry2Ae locus resulting at least initially in greater expression of fitness costs. The recessive nature of fitness costs may explain why the final resistance allele frequencies at both loci were similar despite the fact that the most likely value for the initial Cry1Ab locus frequency was 10-fold greater. Informally, several runs were made without the Cry2Ae toxin to compare the rate of increase of the Cry1Ab resistance allele in a single gene versus a two gene construct, and the mean rate of increase was greater in the single gene case. These results support those of Gould et al. (2006) that not only do dual gene plants increase the expected longevity of toxins but they also increase the delay due to fitness costs.

Based on these results, there may be some criticism about the assumption of recessive inheritance of fitness costs given the lack of extensive genetic data. This assumption is, however, conservative. If fitness costs are additive or even dominant, those fitness costs would be expressed at even lower resistance allele frequencies. The current assumption allowed resistance allele frequencies to increase to higher levels before fitness costs were expressed. The fact that resistance did not evolve in these simulations despite the conservative assumption provides stronger evidence that the risk of resistance evolving over a 30 year time horizon is small.

BPPD response:  The additional modeling information has been reviewed and BPPD concludes that the deficiency has been adequately addressed and resolved. BCS was able to demonstrate that with more conservative (greater) resistance allele frequencies and the inclusion of Cry1Ab corn, TwinLink(TM) corn had a durability of more than 30 years.
 
D) Discussion regarding Deficiency 7:

Deficiency 7: BCS needs to have a well-developed monitoring plan in place before TwinLink(TM) is approved for commercialization (i.e. sampling regions, sampling scheme, etc). BPPD recommends that BCS follow the monitoring paradigm established for other Bt cotton registrations.

BSC response: For TBW and CBW, 10-20 population will be collected from the Cotton Belt; for PBW, 4-6 populations will be collected from the western USA (Arizona, New Mexico, West Texas, and California) during 2011. These samples in conjunction with data collected prior to commercialization will help establish the baseline susceptibility for all target pests. Resistance monitoring will focus on areas of highest adoption and highest risk for TBW and CBW and states where PBW are an economic pest. Where the risk is highest for resistance evolution in the target pests, BCS will collect at least 20 populations. Monitoring reports will include sampling methodology, bioassay methodology, standardized procedures, detection technique and sensitivity, statistical analysis of probability to detect resistance. Discriminating dose assays for TBW, CBW, and PBW will be provided to the Agency on a yearly basis.

BPPD response: BPPD concludes that BCS' monitoring plan reflects the paradigm established for other Bt cotton registrations and the deficiency has been satisfied. However BPPD recommends that in the Terms and Conditions of the registration, BCS be required to submit baseline susceptibility and a diagnostic assay data.

E) Discussion regarding Deficiency 8:

Deficiency 8: BCS needs to have an established remedial action plan in place for each target pest of TwinLink(TM). It is not sufficient to provide an outline or general statement about different kinds of approaches. BPPD recommends that BCS follow the existing remedial action paradigm for Bt cotton PIPs. 

BSC response: For PBW, BSC will use the remedial action plan developed by the Arizona Bt Cotton Working Group. For TBW and CBW, baseline susceptibility data from 2011 and 2012 will define aspects of "suspected" and "confirmed" resistance and field response for each pest. In general, the plan will incorporate the following measures to identify or confirm suspected resistance:

   * Monitoring (as outlined in MRID 47640014);
   * Education of customer with respect to monitoring of unexpected levels of damage, reporting suspected resistance, and applying established remedial insecticide treatments;
   * Confirm resistance when unexpected damage or survival exceed established upper limit of normal baseline levels, LC50/LC99 exceed established upper limit of baseline levels, and resistance is heritable;
   * Alternate methods to control pests as outlined under "Containment" below;
   * Sales restrictions/suspension until data provided and reviewed by EPA support that target pests have regained susceptibility to Bt proteins;
   * Containment through chemical insecticide spraying, accelerated harvest, crop destruction methods, crop rotation, tillage, use of sterile insects, use of products with different modes of action;
   * Monitoring will be increased in affected and surrounding areas; change in type and size of refuge should be considered.

BPPD response: BPPD has reviewed BCS's additional remedial action information and notes that the above is short summary of the applicant's actual response. BPPD noted that the remedial action plans needs to explicitly state that "the Agency is informed within 30 days of confirmed resistance". In addition, BCS should incorporate that "a long-term resistance management action plan will be designed according to characteristics of resistance events and local agronomic practices". If BCS makes these changes, then Deficiency 8 and the Agency's request for more information will have been satisfied. 

Appendix D. III  -  Risk Assessment of data/information to respond to remaining IRM deficiencies. 										
Below is U.S. EPA's review (June 28, 2011) of BCS' resubmissions for FIFRA section 3 application for TwinLink(TM) Cotton expressing a pyramid with Events T304-40 and GHB 119 (MRIDs 48480011, 48480010, and 48495101), EPA File Symbols 264-RNOA and 264-RNOL: PC Codes: 006525 (Cry1Ab), 006600 (Cry 2Ae). In the original IRM risk assessment dated April 22, 2010, EPA identified eight deficiencies, five of which were addressed on 09/28/10. BCS provided additional data/information to satisfy the remaining deficiencies.

BPPD reviewed BCS' submission to address deficiencies noted in the original review of TwinLink(TM) Cotton, a two-gene pyramided PIP targeting Heliothine pests (BPPD 2010). BPPD's conclusions and recommendations are contained in this memo.

CONCLUSIONS AND RECOMMENDATIONS

DOSE VERIFICATION FOR TBW (Deficiency 3)

BCS provided enough data to resolve concerns with the dose for TwinLink(TM) Cotton and its single parental events against TBW, and Deficiency 3 has been addressed in a satisfactory manner. Specifically, based on the field experiment conducted with young cotton plants, BPPD concludes that Cry1Ab and Cry2Ae PIPs express a high dose against Heliothis virescens (TBW) during the early cotton growing season (also late season for Cry2Ae). No larvae on Cry2Ae and Cry1Ab cotton survived or developed to 3[rd] instar during the 11-day test duration, while 2.8% and 4.5% larvae in control treatments reached or surpassed the 3[rd] instar developmental stage. The lab experiment conducted with older cotton plants, however, resulted in 2.17% survival (> 3[rd] instar) on Cry1Ab plants expressing the parental event only suggesting that expression of Cry1Ab was decreased in older plants compared to younger plants. Based on the limited data available at this point, BPPD cautions on the side of error and concludes that Event T304-40 does not express a high-dose against TBW during the late cotton-growing season (based on larval mortality on older cotton plants in the lab). TwinLink(TM) cotton, however, expresses an effective high dose against TBW, and no larval survival was observed in the lab experiment.

MODELING FOR TBW AND TWINLINK(TM) COTTON WITH NATURAL REFUGE (Deficiency 2)
   
BCS provided additional modeling to resolve concerns for their "Natural Refuge" request for TwinLink(TM) Cotton, and Deficiency 2 has been addressed in a satisfactory manner. BPPD concludes that TwinLink(TM) Cotton should not pose a greater risk of resistance evolution in TBW than, for example, Bollgard II Cotton based on the modeling simulations provided in BCS submission. 

Specifically, BCS conducted two modeling analyses (stochastic, spatially explicit and deterministic, spatially implicit) and generally took a conservative approach to modeling the durability of TwinLink(TM) targeting TBW. Mainly, the model landscape accounted for other Cry1A products and model explored several scenarios with increased dominance and lower survival based on the lower protein expression for Cry1Ab cotton based on lab observations. The natural refuge assumptions were based on available natural host information of Mississippi  -  a worst-case candidate state for this IRM strategy. The deterministic simulations estimated that durability of TwinLink(TM) Cotton would be as low as 80 years when all worst-case assumptions were incorporated or as high as 224 years using benchmark assumptions. This represents a 2.8 fold difference between estimated durabilities. The spatially explicit, stochastic simulations estimated that the durability for TwinLink(TM) Cotton would be on average between 20 years (possibly an unrealistic worst-case scenario based on survival percentages used) and 23 years (possibly a reasonable worst-case scenario). Assuming benchmark parameter values for mortality (as used for other Cry1A products in the past), the mean time to estimated product failure (resistance) was 26 years. Whether results differed significantly between any scenarios modeled was unclear since no such analysis and/or explanation accompanied the report. Additionally, comparative analysis between Bollgard II (Monsanto released data used for the natural refuge SAP) and TwinLink(TM) Cotton showed that estimated mean durabilities were comparable.

Although the likelihood for cross-resistance between Cry1Ab and Cry2Ae cotton has been shown to be small, BPPD notes that it would have been more conservative to consider and incorporate a small degree into the modeling analysis of TwinLink(TM) Cotton. Other forms of epistasis (expression of a gene is suppressed by a gene at another locus) were not considered: rather the applicant assumed that survival of genotypes was multiplicative for both loci (the least conservative assumption). The SAP (2011) suggested that other forms of epistasis should be explored for pyramided PIPs to address the risk of resistance evolution (see conclusion section for discussion).

Resistance evolved almost simultaneously to both toxins (under worst case assumptions) when lower protein expression for Cry1Ab was incorporated for older plants and raised some concerns with BPPD. Further analysis should be conducted to assess whether this outcome was an artifact of the modeling design taken by the modeler or a function of real temporal asynchrony from the growing oviposition period between generations and delayed oviposition of adults dispersing from senescing wild hosts to other acceptable hosts. BPPD notes though that the simulations with a lower dominance value for Cry1Ab did not result in simultaneous resistance evolution of the two toxins during the ten replications. It will be important to determine whether this outcome can be replicated for the durability of other pyramided cotton PIPs that express Cry1A Bt toxins which have lower expression levels during the later cotton growing season (resulting in greater dominance values for the target pest). 

BPPD concludes that TwinLink(TM) Cotton deployed with a natural refuge strategy should not be expected to pose a greater risk of resistance evolution in TBW than for example Bollgard II Cotton based on the modeling simulations provided in BCS' submission.

   I. BACKGROUND

TwinLink(TM) cotton is a pyramided transgenic cotton trait that expresses Cry1Ab (Event T304-40), and Cry2Ae proteins (Event GHB119), and a glufosinate ammonium (herbicide) tolerant trait based on LibertyLink technology. TwinLink(TM) was developed by BCS, using conventional breeding techniques. The Cry1 and Cry2 Bt toxin group are effective against lepidoptera and share less than 20% amino acid homology. Crickmore et al. (1998) constructed a phylogenetic tree using amino acid sequence identity as homology criteria for all presently known Bt crystal proteins. This tree indicates that Cry1Ab and Cry2Ae are genetically dissimilar entities and share little sequence homology. 

On April 22[nd], 2010, the BPPD IRM team completed its initial review of BCS' submission for a FIFRA product registration for TwinLink(TM) cotton. During its review, BPPD identified eight IRM deficiencies for which BCS needed to provide clarifications and/or data.

On July 29[th], 2010, BCS submitted their response to BPPD's eight listed IRM deficiencies for TwinLink(TM) (BPPD 2010a). BPPD reviewed BCS' response and additional information and concluded that five of the eight deficiencies were satisfactorily addressed and resolved (BPPD 2010b). The applicant stated that the remaining deficiencies (simulation modeling and second high-dose verification study, both for Tobacco budworm) would be addressed during the next cotton growing season. The third deficiency (#8) related to the remedial action plan and highlighted that BCS needed to explicitly state that "the Agency is informed within 30 days of confirmed resistance". This BPPD review assesses deficiencies 2 and 3 listed below:

      Deficiency 2: In the modeling analysis provided by BCS, only cotton bollworm (CBW) was modeled but not Tobacco budworm (TBW). For a new refuge strategy, such as Natural Refuge, each main target pest needs to be evaluated separately via a simulation model that incorporates biology, genetics, toxicity, and ecology of the target pest. BCS needs to provide simulation modeling for TBW in order to support their natural refuge request.

      Deficiency 3: The natural refuge paradigm was approved for Bollgard II and WideStrike with a high-dose profile for TBW. If BCS wishes to pursue the Natural Refuge strategy for TwinLink(TM), then either a second high-dose verification study will need to be submitted to support that the PIP expresses a high-dose against TBW, or BCS can design a model with their specific dose data for TBW.

   II. BAYER CROPSCIENCE RESUBMISSION MATERIALS

                     1. Field Dose Study (MRID 48480010)

In 2011, a study was conducted in Mississippi cotton fields (MRID 48480010) to estimate mortality of Heliothis virescens (Tobacco budworm, TBW) on caged plants expressing parental Events Cry1Ab and Cry2Ae as well as the pyramided PIP. 

Materials and Methods:

Four treatments were tested: 1) Cry1Ab cotton (T304-40), 2) Cry2Ae cotton (GHB119), 3) TwinLink(TM) cotton (T304-40 x GHB119), and 4) Coker 312 (negative control). A treatment plot consisted of two 38 inch rows of 40 feet in length (0.0029 acres) and each contained approximately 240 plants. Fine mesh netting cages were set up (8 ft x 20 ft) in each plot which covered approximately 120 plants. Artificial infestations were conducted with individuals from a laboratory strain from USDA-ARS, Stonesville (MS). Fifty males and 50 females were released into each cage on two dates separated by one month (two replicates). Mated females oviposited onto cotton plants, and first data collections began four days after introducing the adults into the cages to obtain lower limits of eggs deposited. Ten randomly selected plants in each cage were inspected and scored using binomial decision rule. If a plant contained at least 30 eggs, it was scored with a `yes'. Larval survival was estimated on 20 randomly selected plants within each cage, and survival and developmental stage were recorded. Larvae that survived 11 days of exposure on the plant and had developed to 3[rd] instar were scored as survivors. The lower limit of the proportion of surviving individuals was then computed by assuming that each plant was infested with 30 eggs (actual infestation was higher  -  data not shown here) yielding an initial egg infestation cohort of 600 eggs (20 plants x 30 eggs) per replicate of Bt plant type. 


Results:
One larva developed to 1[st] instar on one Bt plant type (Cry1Ab), while 17 and 37 developed to >=3[rd] instar in the negative control plots (first and second replicate, respectively). Since no larvae survived on the three Bt treatments, there was no need to compute corrected mortality.

Table D.III.1. TBW larval survival after 11 days of exposure to TwinLink(TM) and its parental Events
Treatment Type
                  Number of Eggs (initial cohort  -  minimum)
                           Total Number of Survivors
                              Number of Survivors
                              (>= 3[rd] instar)
                   Proportion of Survivors  -  minimum value
                              (>= 3[rd] instar)
Coker 312 (Control)
    Replicate 1
                                      600
                                      31
                                      17
                                    0.0283
    Replicate 2
                                      600
                                      28
                                      27
                                    0.0450
Cry1Ab  (single Event)
    Replicate 1
                                      600
                                       1
                                       0
                                       0
    Replicate 2
                                      600
                                       0
                                       0
                                       0
Cry2Ae (single Event)
    Replicate 1
                                      600
                                       0
                                       0
                                       0
    Replicate 2
                                      600
                                       0
                                       0
                                       0
Cry1Ab x Cry2Ae (TL)
    Replicate 1
                                      600
                                       0
                                       0
                                       0
    Replicate 2
                                      600
                                       0
                                       0
                                       0
This Table has been modified from the Table in the original submission (p.12, MRID 48480010)

Applicant Conclusions:

"This study provided additional evidence for the high susceptibility of TBW to TwinLink(TM) as well as its parental (single) events. No survivors (>= 3[rd] instar) were found on the parental event, therefore, TwinLink(TM) should provide IRM benefits comparable to or exceeding those of other currently registered pyramided Bt cotton products approved for the natural refuge strategy."

                     2.       Greenhouse Dose Study (MRID 48480011)

In 2011, BCS conducted a greenhouse trial to provide further evidence that TwinLink(TM) Cotton and its parental events express high dose against Heliothis virescens. 

Materials and Methods:

Cotton plants (144-184 days old) were placed into four isolated screened enclosures to hold groups of different Bt phenotypes. The treatments were 1) Cry1Ab cotton (T304-40), 2) Cry2Ae cotton (GHB119), 3) TwinLink(TM) cotton (T304-40 x GHB119), and 4) non-cotton plants (negative control). Each enclosure contained 12 plants of each treatment (three replicates total). Bt and non-Bt plants were artificially infested with 50 and 15 (24-hr old) TBW neonates, respectively. Larval counts were made at day 7 and 14 after infestation. Survival was computed as the proportion of individuals surviving 14 days and completing development to at least 3[rd] instar. Corrected mortality was calculated by dividing `proportional survival on Bt' by `proportional survival on non-Bt cotton'.

Results:

Larval survival on non-Bt and Cry1Ab plants was 69 and five 3[rd] instar larvae out of a total infestation with 540 (on control) and 1800 (on Bt); after correcting for control mortality, survival on Cry1Ab plants was estimated at 0.0217 (see Table 2). On Cry2Ae and TwinLink(TM) cotton plants, no survivors were found (infestation with 1800 larvae).

Table D.III. 5. TBW larval survival on parental Events and pyramided PIP TwinLink(TM) during the late cotton growing season
Treatment
                             Total Plants infested
                           Total Number of Neonates
                           Total Number of Survivors
                           Mean Proportion Survivors
                           Std. Deviation Proportion
                      Corrected Mean Proportion Survivors
Non-Bt
                                      36
                                      540
                                      69
                                     0.128
                                    0.0484
                                       
Cry1Ab
                                      36
                                     1800
                                       5
                                    0.00278
                                    0.00192
                                    0.0217
Cry2Ae
                                      36
                                     1800
                                       0
                                       0
                                       0
                                       0
TwinLink(TM)
                                      36
                                     1800
                                       0
                                       0
                                       0
                                       0
This Table has been modified from the Table in the original submission (p. 11, MRID 484800011

Conclusions:

The level of mortality observed in these trials and levels of survival on single-toxin plants was comparable to or exceeded that of the currently registered Bt cotton products. Lack of survivors on Cry2Ae plants underscores the highly effective dose of this protein in the TwinLink(TM) pyramid. The low level of survival has not been observed under field conditions but could be due to reduced protein expression levels in older plants tested (late-season decline has also been reported for Cry1Ac in currently registered Bt cotton products). This study shows that plants, even late in the production cycle, continue to cause very high mortality to TBW. This should provide additional support for the use of natural refuge strategy for T304-40 x GHB119.

                     3. Simulation Modeling for TwinLink(TM) Cotton and Tobacco Budworm (MRID 48495101)

BCS contracted the simulation modeling to Dr. Mike Caprio from Mississippi State University. Two different models, stochastic and deterministic, were employed to estimate the durability of TwinLink(TM) Cotton in an environment such as the state of Mississippi (worst-case state for natural refuge option).

Materials and Methods:

Spatially explicit, stochastic model: 

This model was based on a previous model by Caprio et al. (2009) for Helicoverpa zea (cotton bollworm, CBW) and was modified to address TBW specifically and to conduct simulations for TwinLink(TM) cotton. Only the modifications to the model were described in BCS' modeling resubmission. A total of six different habitats were modeled (prevalence determined from 2002 data): Bollgard cotton, non-Bt cotton, wild hosts (clover, geranium), and TwinLink(TM) (post-cutout and very late season). Alternate hosts were modeled as separate patches, which in turn forced adult TBW to disperse between patches as wild hosts declined with progressing time. This landscape simulated a situation in the field where alternate hosts would be spatially segregated. All patches were randomly assigned the nature of plant type at the initiation of the simulation. All larvae and pupae were accounted for and placed as pupae in geranium patches at the beginning of each year. Because of complete mixing during the overwintering routine, the habitats were not redistributed at the beginning of each year. This allowed detecting the effects of habitat distribution between simulations.

The dose of Cry1Ab was modeled at 99.9% for the first three generations and subsequently reduced to address the greater survival seen on cotton plants in the lab expressing parental Cry1Ab. Four different levels of decline in protein expression were investigated leading to 0.1% survival (0% decline), 2% survival (level observed in lab), 5% (see Table 3 for genotypic survival), and 10% (extreme case). These levels of decline continued for two more generations until the overwintering subroutine began. Two values of dominance were used: 0.1 (high value) and 0.001 (lower value). Multiplicative toxin interactions were assumed. To overcome extinction of relatively small TBW populations, the fecundity was raised from 50 to 150 eggs/day-female in the model.

Table D.III.3. TBW Survival on TwinLink(TM) Cotton with 5% and 0.1% survival to Cry1Ab and Cry2Ae
Toxin
                                    Cry1Ab
                                    Cry1Ae
Genotypes
                                      SS
                                      RS
                                      RR

Pre-Cutout




SS
                                   0.000001
                                   0.0001009
                                     0.001

RS
                                   0.0001009
                                  0.01018081
                                    0.1009

RR
                                     0.001
                                    0.1009
                                      1.0

Post-Cutout
                                      SS
                                      RS
                                      RR

SS
                                    0.00005
                                   0.000145
                                     0.001

RS
                                    0.00504
                                    0.01463
                                    0.1009

RR
                                     0.05
                                     0.145
                                      1.0
Table modified from original Table (p.11, MRID 48495101)


Deterministic model: 

The Caprio model (based on a construct previously employed by EPA-ORD) was used to model TBW resistance to TL cotton and modified to include refuges of variable size as well as Cry1Ab protein expression decline in older plants. Mortality from Cry2Ae was constant at 99.9% for each generation. For Cry1Ab, mortality was 99.9% for the first three generations (survival of the susceptible genotype was set at 0.0001) and declined thereafter to 98%. The natural refuge percentage varied for each generation and was 20%, 45%, 7%, 1%, 3%, and 3% for each consecutive generation (worst case simulations based on refuge data in worst case county of MS). Fitness for homozygous resistant individuals was 1.0 in all habitats (no fitness cost). Activity of the two toxins was multiplicative (no epistasis). Survivorship of the susceptible genotype was 0.0001% (0.01% x 0.01%). Dominance for TBW was set at 0.001 to compare modeling runs with other models for validation purposes. For further validation, the model was run for a single gene product (Cry1Ac such as for Bollgard), 20% refuge, and 50% reduction in adult emergence due to spraying. The estimated durability was 18.8 years. Given that Bollgard was used for almost 15 years, this would suggest that a dominance value of 0.001 may not be unreasonable and on the high side of realistic. Additionally, comparative modeling was conducted for Bollgard II (Monsanto released data for natural refuge SAP) and TwinLink(TM) to assess the relative durability of TwinLink(TM) to the already registered pyramided PIP.

Results:

Stochastic model: When dominance was set at 0.001, no resistance evolved to TwinLink(TM) within 31 years (including worst-case scenarios with 10% survival after cutout). When dominance was increased to 0.1, results were more complex (see Table 3). With 5% survival in the later season, mean time to resistance was 23.9 years. Greater survival later in the season resulted in faster resistance evolution. Additionally, in almost every simulation resistance to both toxins (Cry1A and Cry2Ae) occurred almost simultaneously. This was unexpected since Cry1Ab moved towards a lower dose later in the growing season, and the expectation was that resistance would evolve to Cry1Ab first. Analysis of data revealed that resistant individuals in two fields became temporarily isolated from the remaining refuge population early in the season. Toxin expression was high at that time and hence, only individuals resistant to both toxins survived. Their homozygous resistant offspring emerged before susceptible individuals dispersed into those fields. This resistance was not caused by isolation as described by Peck et al. (1999) but took place in two separate fields (far enough apart so no dispersal occurred between them) after complete mixing had occurred during the overwintering season (an assumption made by the modeler as described in the methodology section). This apparent temporal asynchrony occurred although no developmental delays were incorporated into the model and was a function of 1) a growing oviposition period between generations (adults emerge over an interval; late emerging cohorts will always be late until redistribution occurs during overwintering season) and 2) adults from senescing wild hosts had to disperse first to a new habitat before oviposition could take place (evidence for this phenomena from shorter "tail" of pupal emergence distribution from wild hosts compared to distributions from TL cotton). The data suggest that resistance evolved in "the tails of the generational distributions". Resistance observed under these model parameters was not specific to TwinLink(TM) Cotton but could be generated by a hypothetical two-gene PIP consisting of two high dose toxins (99.9% mortality). The modeler contended that the result might be an artifact based on model design aimed at addressing TBW biology. Further research, independent of any particular PIP, may be warranted. 

The model was run using Monsanto's parameter values and dominance values of 0.001 and 0.1. With lower dominance, resistance did not evolve in any of the 10 replicates; with lower dominance values, resistance evolved as early as in 10.1 years and mean time to resistance was 27.3 years (comparable to TwinLink(TM) results).

Table D.III.6. Mean time to resistance for TBW to TwinLink(TM) Cotton 
                        Survival to Cry1Ab Post-Cutout
                         Mean Time to Resistance (yrs)
                             Number of Simulations
                                    0.1%[1]
                                     26.0
                                      10
                                     2%[2]
                                     23.5
                                      10
                                      5%
                                     23.9
                                      10
                                      10%
                                     20.6
                                      10
Survival to Cry2Ae was held constant during the different simulations
[1] default survival value used by registrants based on field data from Cry1A plants prior to cutout.
[2] Survival value observed by BCS in 2011 greenhouse studies of plants after cutout.
Table modified from original Table (p.12, MRID 48495101)

Deterministic Model: The deterministic runs of the model resulted in an estimated 224 years before resistance evolved in TBW to TwinLink(TM) (both toxins) under benchmark assumptions. When the dominance value was increased to 0.01 (for later generations of TBW in response to lower expression levels in older cotton plants), it took 80 years before resistance evolved to TwinLink(TM) (see Table D.III.5 below). Unlike other previous models for TBW and currently registered Bt cotton PIPs expressing Cry1A protein, BCS incorporated a seasonal decline in Cry1Ab protein expression, which somewhat elevated the risk of resistance evolution.

Table D.III.5. Durability of TwinLink(TM) Cotton based on the deterministic model simulations
                          Dominance Value for Cry1Ab
                           Time to Resistance (yrs)
                                     0.001
                                      224
                                    0.01[1]
                                      80
TL simulations run with parameter values as close to Monsanto's values for Bollgard II
[1] Dominance is 10x that of Monsanto's value




Conclusions:

"Based on both stochastic and deterministic models, using worst-case scenarios for refuge size and contrasts of other key model parameters, TwinLink(TM) would not be expected to change the risk of resistance over existing dual-Bt products. Risk of resistance to TwinLink(TM) and all other Cry1A-based dual-Bt products registered in the US may be somewhat higher than indicated by previous simulations, owing to possible late-season decline in Cry1Ac expression."

   III. BPPD REVIEW AND CONCLUSIONS

DOSE (Deficiency 3): Based on the field experiment conducted with young cotton plants, BPPD concludes that Cry1Ab and Cry2Ae (individual parental events) express a high dose against Heliothis virescens (TBW) during the early cotton growing season (also late season for Cry2Ae). No larvae on Cry2Ae and Cry1Ab survived or developed to 3[rd] instar during the 11-day test duration, while 2.8% and 4.5% larvae in control treatments reached or surpassed the 3[rd] instar developmental stage. The lab experiment conducted with older cotton plants, however, resulted in 2.17% survival (> 3[rd] instar) on Cry1Ab plants expressing the parental event only suggesting that expression of Cry1Ab protein was decreased in older plants compared to younger plants. Based on the limited data available at this point, BPPD cautions on the side of error and concludes that Event T304-40 does not express a high-dose against TBW during the late cotton-growing season (based on larval mortality on older cotton plants in the lab). TwinLink(TM) cotton, however, expresses an effective high dose against TBW, and no larval survival was observed in the lab experiment.

MODELING (Deficiency 2): BCS conducted two modeling analyses (stochastic spatially explicit and deterministic, spatially implicit) and generally took a conservative approach to modeling the durability of TwinLink(TM) Cotton targeting TBW. Dr. Mike Caprio from Mississippi State University incorporated other Cry1A products into the landscape and explored several scenarios with increased dominance and lower survival based on the lower protein expression observed in the lab for Cry1Ab cotton. The natural refuge assumptions in the model were based on available host data of Mississippi - a worst-case candidate state for this IRM strategy. The deterministic simulations estimated that durability of TwinLink(TM) Cotton would be as low as 80 years when all worst-case assumptions were incorporated or as high as 224 years using benchmark assumptions. This was a 2.8 fold difference between estimated durabilities. The spatially explicit, stochastic simulations estimated that the durability for TwinLink(TM) Cotton was on average between 20 years (probably an unrealistic worst-case scenario based on survival percentages used) and 23 years (probably reasonable worst-case scenario). Assuming benchmark parameter values for mortality (as used for other Cry1A products in the past), the mean time to estimated product failure (resistance) was 26 years. Whether results differed significantly between any scenarios modeled was unclear since no such analysis and/or explanation accompanied the report. Additionally, comparative analysis between Bollgard II (Monsanto released data 	presented to SAP for Natural Refuge assessment) and TwinLink(TM) Cotton showed that estimated mean durabilities were comparable between the two pyramided PIPs.

Although the likelihood for cross-resistance between Cry1Ab and Cry2Ae cotton has been shown to be small, BPPD notes that it would have been more conservative to consider and incorporate a small degree of epistasis (expression of a gene is suppressed by a gene at another locus) into the modeling analysis of Cotton. Other forms of epistasis were not considered: rather the applicant assumed that survival of genotypes was multiplicative for both loci (the least conservative assumption). The SAP (2011) suggested that additional forms of epistasis should be explored for pyramided PIPs such as, for example, developmentally restricted expression of low levels of Cry-protease and where resistance at the a locus could be determined by the most rapidly evolving locus, which could drag other resistance loci along and, thereby, increase the rate of resistance evolution beyond what would be estimated by a purely multiplicative approach.

The discussion as to why resistance evolved almost simultaneously to both toxins (with assumption of increased dominance) raised some concerns with BPPD. Further analysis should be conducted to assess whether this outcome was an artifact of the modeling design taken by the modeler or a function of real temporal asynchrony from the growing oviposition period between generations and delayed oviposition of adults dispersing from senescing wild hosts to other acceptable hosts. BPPD notes though that the simulations with a lower dominance value for Cry1Ab did not result in simultaneous resistance evolution of the two toxins during the ten replications. It will be important to determine whether this outcome can be replicated for the durability of other pyramided cotton PIPs that express Cry1A Bt toxins with lower expression levels during the later cotton growing season (resulting in greater dominance values for the target pest). 

BPPD concludes that TwinLink(TM) Cotton deployed with a natural refuge strategy should not be expected to pose a greater risk of resistance evolution in TBW than, for example, Bollgard II Cotton based on the modeling simulations provided in BCS submission.


APPENDIX E: References 

Insect Resistance Management

1. BPPD, 2010a. EPA review of BCS' insect resistance management plan for a section(3) full commercial registration request for TwinLink(TM) Cotton (EPA File Symbol 264-RNOU, MRIDs 4760014 and 47634823). Memorandum from JC Martinez to S. Bacchus and D. Greenway on April 4.

2. BPPD, 2010b. EPA Review of Bayer's Response to IRM Deficiencies noted in BPPD's risk assessment for TwinLink(TM) dated 4/22/2010 (EPA File Symbol 264-RNOU, MRIDs 4760014 and 47634823). Memorandum from JC Martinez to S. Bacchus and D. Greenway on July 29.

3. Caprio MA, CD Parker, and JC Schneider. 2009. Future fitness of female insect pests in temporally stable and unstable habitats and its impact on habitat utility as refuges for insect resistance management. Journal of Insect Science, Vol. 9(44): 1-10

4. MA Caprio. 2011. A risk assessment of TwinLink(TM) Cotton and Tobacco Budworm (Heliothis virescens) in the Midsouth. Bayer CropScience unpublished report submitted to the Agency, MRID 48495101.

5. Crickmore  N, DR Zeigler, J Feitelson, E Schnepf, D Van Rie, D Lereclus, J Baum , and DH Dean. 1998. Revisions of the Nomenclature for the Bacillus thuringiensis pesticidal crystal proteins. Microbiology and Molecular Biology Reviews, Vol. 62 (3): 807-813.

6. EPA Scientific Advisory Panel (SAP). 2011. Transmittal Meeting Minutes of the FIFRA Scientific Advisory Panel Meeting Held December 8-9, 2010 to Address Scientific Issues Associated with Insect Resistance Management for SmartStax(TM) Refuge-in-the-Bag, a Plant-Incorporated Protectant (PIP) Corn Seed Blend. Memorandum from Dr. Sharlene Matten (DFO) to Dr. Steven Bradbury (Director of OPP) on March 3, 2011.

7. Navacerrad JJ, J Pierson, I Coats, and PL Hunst. 2011. High dose study: estimating the mortality of Heliothis virescens in a greenhouse on plants on T304-40 x GHB119 (TwinLink(TM)) Cotton (BCS-GH004-7 x BCS-GH005-8) and parental Event plants T304-40 (Cry1Ab) and GHB119 (Cry2Ae). Bayer CropScience unpublished report submitted to the Agency, MRID 48480011

8. W Mullins. 2011. High dose study: estimating the mortality of Heliothis virescens in a Mississippi cotton field on caged plants on T304-40 x GHB119 (TwinLink(TM)) Cotton (BCS-GH004-7 x BCS-GH005-8) and parental Event plants T304-40 (Cry1Ab) and GHB119 (Cry2Ae). Bayer Crop Science unpublished report submitted to the Agency, MRID 48480010

9. Peck SL, F Gould, and SP Ellner. 1999. Spread of resistance in spatially-extended regions of transgenic cotton: implications for management of Heliothis virescens (Lepidoptera: Noctuidea). Journal of Economic Entomology, Vol. 92: 1-1
REFERENCES

Other References:

U.S. EPA. 1998. FIFRA Scientific Advisory Panel Subpanel on Bacillus thuringiensis (Bt) Plant-Pesticides and Resistance Management, February 9 and 10, 1998

U.S. EPA. 2000. FIFRA Scientific Advisory Panel Subpanel on Bt Plant-Pesticides Risk and Benefit Assessments: Insect Resistance Management, October 18-20, 2000

U.S. EPA. 2001. Biopesticides Registration Action Document  -  Bacillus thuringiensis Plant Incorporated Protectants, http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad.htm

U.S. EPA. BPPD. 2009. Review of 2007 monitoring data submitted by ABSTC for lepidopteran Bt corn registrants (MRID 47413901). Memorandum from S. Borges to M. Mendelsohn dated April 8[th], 2009.


U.S. EPA REVIEWS  -  Insect Resistance Management

U.S. EPA. 2010a. Memorandum J. Martinez and A. Reynolds. BCS' Insect Resistance Management Plan for a Section (3) Full Commercial Registration Request of TwinLink Cotton (EPA Reg. No. 264-RNOU, MRIDs 4760014 and 47634823). April 22.

U.S. EPA. 2010b. Memorandum. J. Martinez and A. Reynolds. September 28, 2010. EPA Review of Bayer's Response to IRM Deficiencies noted in BPPD's risk assessment for TwinLink dated 4/22/2010 (EPA File Symbols 264-RNOU, 264-RNOL, and 264-RNOA, MRIDs 4760014 and 47634823). September 28. 

U.S. EPA. 2011. Memorandum. J. Martinez and A. Reynolds. EPA review of Bayer CropScience's resubmissions for TwinLink(TM) Cotton sec(3) application expressing a pyramid with Events T304-40 and GHB 119 (MRIDs 48480011, 48480010, and 48495101), EPA File Symbol 264-RNOA, -RNOL: PC Codes: 006525 (Cry1Ab), 006600 (Cry 2Ae). June 28.
 

Appendix E.1. Bibliography of data submitted for registration of Bt Cry1Ab (Event T304-40) and TwinLink(TM) Cotton, its combination with Bt Cry2Ae (Event GHB119).
Most of the submissions below refer to Cry2Ae, and some of them also refer to TwinLink(TM). There may be repeats of those made for Cry2Ae in the Bibliography cited for Cry1Ab because they were submitted for TwinLink(TM), which contains both proteins.

I.	Cry 1Ab and TwinLink(TM) Cotton - Data Submissions 
                                     MRID
                     Citation Reference  (Bt Cry 1Ab PIP)
47076900
Bayer Cropscience LP. (2007) Submission of Toxicity Data in Support of the Experimental Use of Bacillus thuringiensis Cry1Ab and Bacillus thuringiensis Cry2Ae for Use on Cotton. Transmittal of 2 Studies. 
47076902
Rouque, D. (2006) Cry2Ae (GEM2) Protein: Acute Toxicity by Oral Gavage in Mice. Project Number: SA/06235, BBS06/010. Unpublished study prepared by Bayer Cropscience. 62 p.
47171700
Bayer Cropscience (2007) Submission of Residue Data in Support of the Experimental Use of Bacillus thuringiensis Cry1Ab Insecticidal Protein on Cotton Plants. Transmittal of 1 Study.
47171701
Hayes, R. (2007) Final Field Report for 2006 Season of Bacillus thuringiensis Cry1Ab Insecticidal Protein as Expressed in Cotton Plants. Project Number: 264/EUP/140. Unpublished study prepared by Bayer Cropscience. 6 p.
47634800
Bayer CropScience (2008) Submission of Efficacy Data in Support of the Application for Registration of BCS Cry1Ab Cotton Event T304-40. Transmittal of 20 Studies.
47634801
Ferullo, J. (2008) Product Characterization of Cry1Ab Cotton Event T304-40. Unpublished study prepared by Bayer BioScience N.V. 22 p.
47634802
Nennstiel, D. (2008) Product Characterization: Nutritional Characterization of Cry1Ab Cotton Event T304-40. Unpublished study prepared by Bayer CropScience LP. 8 p.
47634803
Currier, T.; Massengill, J. (2008) Protein Expression Analysis of Cotton Event T304-40, Expressing Cry1Ab and PAT/bar Proteins. Project Number: CP07B005. Unpublished study prepared by Bayer CropScience LP. 158 p.
47634804
Martone, A. (2008) Analyses of Raw Agricultural Commodity (Fuzzy Seed) of Cry1Ab Cotton Event T304-40 for PAT/bar and Cry1Ab and its Non-Transgenic Counterpart for PAT/bar and Cry1Ab Proteins. Project Number: CP07B011. Unpublished study prepared by Bayer CropScience LP. 154 p.
47634805
Martone, A. (2008) Structural and Functional Equivalence of Cry1Ab and PAT/bar Proteins Produced in Escherichia coli to Cry1Ab and PAT/bar Proteins from Event T304-40 and TwinLink(TM) Cotton, Gossypium hirsutum. Project Number: CP08B003. Unpublished study prepared by Bayer CropScience LP. 52 p.
47634806
Nennstiel, D. (2008) ORF Analysis - Toxicology (Human Health Assessment). Unpublished study prepared by Bayer Cropscience. 10 p.
47634809
Stephenson, I. (2008) Cry1Ab Protein: Epitope Homology, N-glycosylation and Overall Amino Acid Sequence Homology Search with Known Toxins and Allergens. Supplement to MRID 46708802 and MRID 46708803. Unpublished study prepared by Bayer BioScience N.V. 6 p.
47634812
Nennstiel, D. (2008) Toxicology (Human Health Assessment): History of Safe Use. Unpublished study prepared by Bayer BioScience N.V. 49 p.
47634813
Chalmers, A. (2008) Summary of Non-Target Organism Testing and Assessment of Risk of Gossypium hirsutum: Transformation Event T304-40 Expressing Cry1Ab Protein. Project Number: ECORA1. Unpublished study prepared by Bayer CropScience. 67 p.
47634815
Patnaude, M. (2008) Laboratory Study to Determine the Effects of Cry1Ab Protein on the Predatory Beetle Coleomegilla maculata. Project Number: EB99L008, 13798/6217. Unpublished study prepared by Springborn Smithers Laboratories. 60 p.
47634816
Scott, A. (2008) Cry1Ab Protein: Toxicity to Green Lacewing (Chrysoperla rufilabris). Project Number: EB99L009. Unpublished study prepared by Bayer CropScience LP. 4 p.
47634818
Bushey, D. (2008) Request for Waivers from the Requirement to Conduct a Non-Target Soil Invertebrate Study: Earthworm. Unpublished study prepared by Bayer CropScience LP. 6 p.
47634819
Sayers, L. (2008) Cry1Ab Protein: Ten Day Toxicity Test to Water Fleas (Daphnia magna) Under Static-Renewal Conditions. Project Number: 13798/6232, EB99L010. Unpublished study prepared by Springborn Smithers Laboratories. 45 p.
47634820
Martone, A. (2008) The Use of an Insect Heliothis virescens Bioassay to Determine the DT50 of the Cry1Ab Protein Produced from Escherichia coli After Aerobic Soil Degradation. Project Number: CP07B010. Unpublished study prepared by Bayer CropScience LP. 152 p.
47634821
Jesudason, P. (2008) Cry1Ab: Efficacy Assessment (Field and Laboratory). Unpublished study prepared by Bayer CropScience LP. 5 p.
47634822
Bushey, D. (2008) Request for a Waiver from the Requirement to Prepare a Public Interest Document. Unpublished study prepared by Bayer CropScience LP. 6 p.
47634823
Bushey, D. (2008) Request for a Waiver from the Requirement to Develop an Insect Resistance Management Plan. Unpublished study prepared by Bayer CropScience LP. 6 p.
47634824
Nennstiel, D. (2008) Detection Methods: rtPCR. Unpublished study prepared by Bayer BioScience N.V. 6 p.
47634825
Nennstiel, D. (2008) Protein Detection Methods: Lateral Flow Strip (LFS). Unpublished study prepared by EnviroLogix, Inc. 8 p.
47634826
Nennstiel, D. (2008) Protein Detection Methods: Enzyme Linked Immuno Sorbent Assay (ELISA). Unpublished study prepared by Strategic Diagnostics, Inc. and EnviroLogix, Inc. 18 p.
47640000
Bayer Cropscience LP (2008) Submission of Product Chemistry, Efficacy, Toxicity, Exposure and Risk and Public Interest Data in Support of the Application for Registration of BCS TwinLink(TM) Cotton Event T304-40 x GHB 119 Cotton. Transmittal of 14 Studies.
47640001
Ferullo J. (2008) TwinLink(TM) Product Characterization - Molecular Characterization. Unpublished study prepared by Bayer BioScience N.V. 6 p.
47640002
Jesudason, P. (2008) Product Characterization - Nutritional Characterization of Twinlink. Unpublished study prepared by Bayer Cropscience LP. 8 p.
47640003
Mackie, S. (2008) Protein Expression Analysis for T304-40, GHB119 and TwinLink(TM) Transgenic Cotton. Project Number: CP08B002, BBS07/013, BB06/010. Unpublished study prepared by Bayer Cropscience LP. 155 p.
47640004
Martone, A. (2008) Analyses of Raw Agricultural Commodity (Fuzzy Seed) of TwinLink(TM) Cotton for PAT/bar, Cry2Ae and Cry 1Ab and its Non-Transgenic Counterpart for PAT/bar, Cry2Ae and Cry1Ab Proteins. Project Number: CY07B008, BTS/0066/06, BBS07/006. Unpublished study prepared by Bayer Cropscience LP. 216 p.
47640005
Martone, A. (2008) Structural and Functional Equivalence of Cry1Ab, Cry2Ae and PAT/bar Proteins Produce in Escherichia coli and Bacillus thuringiensis to the Cry1Ab, Cry2Ae and PAT/bar Proteins from Events T304-40, GHB119 and TwinLink(TM) Cotton Seed, Gossypium hirsutum. USA, 2008. Project Number: CP08B011, BTS/0077/08, BBS07/006. Unpublished study prepared by Bayer Cropscience LP. 70 p.
47640006
Martone, A. (2008) Structural and Functional Equivalence of Cry1Ab, Cry2Ae and PAT/bar Proteins Produced in Bacillus thuringiensis (Bt) and Escherichia coli to Cry1Ab, Cry2Ae and PAT/bar in TwinLink(TM) Cotton, Gossypium hirsutum, USA, 2007. Project Number: CY07B006, BBS07/013, BIOX/001/PROTSUPP/07. Unpublished study prepared by Bayer Cropscience LP. 402 p.
47640007
Bushey, D. (2008) Toxicology: Summary Human Health Assessment. Unpublished study prepared by Bayer BioScience N.V. 8 p.
47640008
Chalmers, A. (2008) Summary of Non-Target Organism Testing and Assessment of Risk of TwinLink(TM) Cotton Expressing Cry1Ab and Cry2Ae Proteins. Project Number: ECORA3. Unpublished study prepared by Bayer Cropscience LP. 17 p.
47640009
Richards, K. (2008) Evaluation of the Dietary Effect(s) of Pollen from TwinLink(TM) on Honey Bee Larvae (Apis mellifera L.) Development. Project Number: EB99X037, CAR/204/08, BTS/0081/08. Unpublished study prepared by California Agricultural Research, Inc. and Bayer CropScience Ecotoxicology Department. 48 p.
47640010
Patnaude, M. (2009) Chronic Toxicity to Collembola (Folsomia candida) Using Transgenic Lyophilized TwinLink(TM) Cotton Leaf. Project Number: EB99X036, 13798/6223. Unpublished study prepared by Springborn Smithers Laboratories. 40 p.
47640011
Stafford, J. (2008) Broiler Chicken Nutritional Equivalency Study with TwinLink(TM) Cotton. Project Number: 13798/4118, TX99X106, M/310302/01/1. Unpublished study prepared by Springborn Smithers Laboratories. 223 p.
47640012
Rinehardt, M.; Holloway, J. ; Bushey, D. (2008) Agronomic Performance and Efficacy of TwinLink(TM) Cotton Obtained by Conventional Breeding of Cry1Ab Cotton Event T304-40 and Cry2Ae Cotton Event GHB119: 2007-2008 USA Production Seasons. Project Number: 07/044/102N, IR/7D/GH/MR, 07/059104N. Unpublished study prepared by Bayer Cropscience LP. 68 p.
47640013
Maclntosh, S.; Bushey, D.; Christian, M.; et al. (2008) Public Interest Document for Bacillus thuringiensis Cry1Ab and Cry2Ae Insecticidal Proteins as Expressed in Combined Trait Cotton (TwinLink(TM) Cotton) Plants. Unpublished study prepared by Bayer CropScience LP. 36 p.
47640014
Maclntosh, S.; Holloway, J.; Jansens, S.; et al. (2008) Insect Resistance Management Plan: TwinLink(TM) Cotton. Unpublished study prepared by Bayer CropScience LP. 88 p.
47651100
Bayer Cropscience LP - Bioscience (2008) Submission of Fate and Toxicity Data in Support of the Application for Registration of BCS Cry1Ab Cotton Event T304-40. Transmittal of 6 Studies.
47651101
Rouquie, D. (2007) Cry1Ab Protein: In Vitro Digestibility Study in Simulated Intestinal Fluid. Project Number: M/295264/01/1, SA/07111, LYNX/PSI/N/TX99L035. Unpublished study prepared by Bayer Cropscience. 54 p.
47651102
Rouquie, D. (2007) Cry1Ab Protein: In Vitro Digestibility Study in Human Simulated Gastric Fluid. Project Number: M/295272/01/1, SA/07110, LYNX/PSIN/TX51X006. Unpublished study prepared by Bayer Cropscience. 56 p.
47651103
Rouquie, D. (2007) Cry1Ab Protein: Acute Toxicity by Oral Gavage in Mice. Project Number: M/295507/01/1, SA/07109, LYNX/PSI/NTX51X007. Unpublished study prepared by Bayer Cropscience. 58 p.
47651104
Rouquie, D. (2007) Cry1Ab Protein: Heat Stability Study. Project Number: M/295690/01/1, SA/07112, LYNX/PSI/N/TX99L034. Unpublished study prepared by Bayer Cropscience. 51 p.
47651105
Richards, K. (2008) Evaluation of the Dietary Effect(s) of a Cry1Ab Protein on Honey Bee Larvae (Apis mellifera L.). Project Number: EB99X030, CAR/177/07, BBS07/006. Unpublished study prepared by California Agricultural Research, Inc. 111 p.
47651106
Patnaude, M. (2008) Chronic Toxicity to Collembola (Folsomia candida) Using Cry1Ab Protein. Project Number: 13798/6218, EB99X028, BBS07/006. Unpublished study prepared by Springborn Smithers Laboratories. 61 p.
47692700
Bayer CropScience LP (2009) Submission of Toxicity Data in Support of the Application for Registration of BCS Cry1Ab Cotton Event T304-40. Transmittal of 1 Study.
47692701
Patnaude, M. (2009) Cry1Ab Protein - Toxicity to Green Lacewing (Chrysoperla rufilabris) Following OPPTS Guideline 885.4340. Project Number: 13798/6230, EB99L009, 082808. Unpublished study prepared by Springborn Smithers Laboratories. 48 p.
47850300
Bayer CropScience, LP (2009) Submission of Environmental Fate Data in Support of the Experimental Use of TwinLink(TM) Cotton on Cotton Plants. Transmittal of 1 Study.
47850301
Coats, I. (2009) TwinLink(TM) Cotton (BCS-GH004-7 x BCS-GH005-8): Investigation of Pollen Flow in Puerto Rico. Unpublished study prepared by Bayer CropScience, LP. 24 p.
47871400
Bayer CropSciences (2009) Submission of Product Chemistry, Environmental Fate and Toxicity Data in Support of the Application for Registration of BCS Cry1Ab Cotton Event T304-40. Transmittal of 4 Studies. 
47871401
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Product Analysis Studies as Outlined in 40 CFR 158.2120-Cry1Ab. Unpublished study prepared by Bayer CropScience LP. 7 p.
47871402
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Toxicology Studies to Test for Health Effects as Outlined in 40 CFR 158.2140-Cry1Ab. Unpublished study prepared by Bayer CropScience LP. 9 p.
47871403
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Nontarget and Environmental Fate Studies to Test for Ecological Effects as Outlined in 40CFR 158.21-Cry1Ab. Unpublished study prepared by Bayer CropScience LP. 7 p.
47871404
Hunst, P. (2009) Bridging Data from TwinLink(TM) Cotton Avian Nutritional Study to Parental Events T304-40 and GHB119. Unpublished study prepared by Bayer CropScience LP. 7 p.
47871600
Bayer CropScience, LP (2009) Submission of Product Chemistry and Toxicity Data in Support of the Application for Registration of BCS TwinLink(TM) Cotton Event T304-40xGHB 119 Cotton. Transmittal of 6 Studies. 
47871601
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Product Analysis Studies. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47871602
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Toxicology Studies to Test for Health Effects as Outlined in TwinLink(TM). Unpublished study prepared by Bayer CropScience. 9 p.
47871603
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Nontarget and Environmental Fate Studies to Test for Ecological Effects - TwinLink(TM). Unpublished studies prepared by Bayer CropScience. 7 p.
47871604
Currier, T. (2009) Protein Expression Analysis for Pat/bar, Cry1Ab and Cry2Ae Proteins in Whole TwinLink(TM) Plants Grown in the Field. Project Number: CP09Q005. Unpublished study prepared by Bayer CropScience, LP. 28 p.
47871605
Hunst, P. (2009) Investigation of Synergistic Effects of the Cry1Ab and Cry2Ae Proteins expressed in TwinLink(TM) Cotton (BCS-GH004-7 x BCS-GH005-8) on Target Insect Pests Using Leaf Bioassay. Unpublished study prepared by Bayer CropScience, LP. 8 p.
47871606
Coats, I. (2009) Supplement to MRID # 47640014: Insect Resistance Management Plan: TwinLink(TM) Cotton. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47903100
Bayer CropScience LP (2009) Submission of Toxicity Data in Support of the Application for Registration of BCS TwinLink(TM) Cotton Event T304-40 x GHB 119 Cotton. Transmittal of 1 Study.
47903101
Hunst, P. (2009) Request for Waiver from the Requirement to Conduct an Avian Oral Toxicity Test: TwinLink(TM). Unpublished study prepared by Bayer CropScience LP. 16 p.
47903200
Bayer CropScience LP (2009) Submission of Toxicity Data in Support of the Application for Registration of BCS CRY1AB Cotton Event T304-40. Transmittal of 1 Study.
47903201
Hunst, P. (2009) Request for Waiver from the Requirement to Conduct an Avian Oral Toxicity Test: TwinLink. Unpublished study prepared by Bayer CropScience LP. 16 p.
47928800
Bayer CropScience, LP-BioScience. (2009) Submission of Efficacy Data in Support of the Registration of BSC TwinLink(TM) Cotton Event T304-40 X GHB 119 Cotton. Transmittal of 1 Study.
47928801
Coats, I. (2009) Addendum to Product Chemistry for TwinLink: Structural and Functional Equivalence of Cry1Ab and Cry2Ae and PAT/bar Proteins Produced in Bacillus thuringiensis (Bt) and Escherichia coli to Cry1Ab, Cry2Aa and PAT/Bar in TwinLink(TM) Cotton, Gossypium hirsutum, USA, 2007: Final Report. Project Number: CY07B006. Unpublished study prepared by Bayer CropScience, LP. 7 p.
48166300
Bayer CropScience LP-BioScience (2010) Submission of Efficacy Data in Support of the Experimental Use of Bacillus thuringiensis Cry2Ae Insecticidal Protein and Bacillus thuringiensis Cry1Ab and Cry2Ae Insecticidal Proteins on Cottons. Transmittal of 1 Study.
48166301
Bushey, D.; Acker, T.; Price, K.; et al. (2010) TwinLink(TM) Cotton (BCS-GH004-7 x BCS-GH005-8): Investigation of Pollen Flow in Puerto Rico: Year 2: 2009-2010. Project Number: M/386577/01/1. Unpublished study prepared by Bayer CropScience. 21 p.
48427900
Bayer CropScience (2011) Submission of Efficacy Data in Support of the Experimental Use of T304-40 x GHB119 x COT102 x COT67B in Cotton. Transmittal of 1 Study.
48427901
Robinson, T. (2011) Cry1Ab, Cry2Ae, PAT/bar, FLCry1Ab, Vip3Aa, APH4 Levels in Selected Tissues of the Combined Trait Product T304-40 x GHB119 x COT102 x COT67B (BCS-GHOO4-7 x BCS- GH005-8 x SYN-IR1O2-7 x SYN-IRO67-1). Project Number: 10/BM45L001. Unpublished study prepared by Bayer CropScience, LP. 22 p.
48471900
Bayer CropScience (2011) Submission of Product Chemistry Data in Support of the Experimental Use of Cry2Ae (GEM2) in Cotton. Transmittal of 1 Study.
48471901
Currier, T. (2011) Analysis to Determine if the Cry2Ae (GEM2) Protein from Cotton Leaves is Glycosylated: Final Report. Project Number: BK05B003. Unpublished study prepared by Bayer CropScience. 32 p.
48480000
Bayer CropScience LP (2011) Submission of Product Chemistry and Efficacy Data in Support of the Applications for Registration of BCS Cry1Ab Cotton Event T304-40, BCS Cry2Ae Cotton Event GHB119 and BCS TwinLink(TM) Cotton Event T304-40 x GHB 119 Cotton, and the Petition for Tolerance of Cry2Ae. Transmittal of 11 Studies.
48480001
Habex, V.; Verhaeghe, S. (2008) Detailed Insert Characterization of Gossypium hirsutum Transformation Event GHB119 by Southern Blot Analysis. Project Number: BBS07/009, M/308404/02/1. Unpublished study prepared by Bayer BioScience N.V. 43 p.
48480002
Moens, S. (2008) Confirmation of the Absence of Vector Backbone Sequences in Gossypium hirsutum Transformation Event GHB119. Project Number: BBS06/007, M/311127/02/1. Unpublished study prepared by Bayer BioScience N.V. 40 p.
48480003
Moens, S.; Criel, I. (2008) Detailed Insert Characterization of Gossypium hirsutum Transformation Event T304-40. Project Number: BBS07/014, M/311245/03/1. Unpublished study prepared by Bayer BioScience N.V. 62 p.
48480004
Habex, V.; Van Herck, H. (2007) Confirmation of the Absence/Presence of Vector Backbone Sequences in Gossypium hirsutum Transformation Event T304-40: Final Report. Project Number: BBS05/003, M/308400/02/1. Unpublished study prepared by Bayer BioScience N.V. 41 p.
48480005
Currier, T. (2011) Characterization of Cry1Ab and (Inert Ingredient) Proteins Produced in Event T304-40 Cotton, Gossypium hirsutum USA, 2011. Project Number: 10/BM99L508, M/406631/01/1. Unpublished study prepared by Bayer CropScience LP. 31 p.
48480006
Moens, S. (2011) Characterization of Cry2Ae and (Inert Ingredient) Proteins Expressed in Gossypium hirsutum Transformation Event GHB119. Project Number: BIO2/005/PROTEQUL/512, M/406615/02/1. Unpublished study prepared by Bayer BioScience N.V. 35 p.
48480007
Currier, T. (2011) Structural Stability Analysis of Gossypium hirsutum Combined Event TwinLink(TM) X GlyTol (GHB614): Final Report. Project Number: CP08Q012/2. Unpublished study prepared by Bayer CropScience LP. 34 p.
48480008
Currier, T.; Harbin, A. (2011) Protein Expression Analysis for GlyTol x TwinLink(TM) (GHB614 x T304-40 x GHB119) Cotton Grown in the Field, USA 2010. Project Number: 10BM99L440B, M/406245/01/1. Unpublished study prepared by Bayer CropScience LP. 125 p.
48480009
Bushey, D.; Jansens, S.; Becker, A. (2010) Insect Bioassay of Plant Material Expressing Cotton Events T304-40, GHB119, and TwinLink(TM) (T304-40xGHB119). Project Number: BIO2/005, INSECT/BIOASSAY/FROM/MONHEIM. Unpublished study prepared by Bayer CropScience. 22 p.
48480010
Mullins, W. (2011) High Dose Study: Estimating the Mortality of Heliothis virescens in a Mississippi Cotton Field on Caged Plants of T304-40 x GHB119 (TwinLink(TM)) Cotton (BCS-GHOO4-7 x BCS-GHOO5-8) and Parental Events T304-40 (Cry1Ab) and GHB119 (Cry2Ae). Unpublished study prepared by Bayer CropScience LP. 16 p.
48480011
Dennehy, T. (2011) High Dose Study: Estimating the Mortality of Heliothis virescens in a Greenhouse on Plants of T304-40 x GHB119 (TwinLink(TM)) Cotton (BCS-GHOO4-7 x BCS-GHOO5-8) and Parental Event Plants T304-40 (Cry1Ab) and GHB119 (Cry2Ae). Project Number: BA44Y001. Unpublished study prepared by Bayer CropScience LP. 16 p.
48495100
Bayer CropScience (2011) Submission of Efficacy Data in Support of the Application for Registration of the TwinLink(TM). Transmittal of 1 Study.
48495101
Dennehy, T. (2011) A Risk Assessment of TwinLink(TM) Cotton and Tobacco Budworm (Heliothis virescens) in the Midsouth. Unpublished study prepared by Mississippi State University. 15 p.


Appendix E.2. Bibliography of data submitted for Bacillus thuringiensis Bt Cry 2Ae (Event GHB119) and TwinLink (TM), its combination with Bt Cry1Ab (Event T304-40)  

Most of the submissions below refer to Cry2Ae, and some of them also refer to TwinLink(TM) Cotton. There may be repeats of those made for Cry2Ae in the Bibliography cited for Cry1Ab because they were submitted for TwinLink(TM) Cotton, which contains both proteins.

                                     MRID
                      Citation Reference (Bt Cry 2Ae PIP)
46708900
Bayer Cropscience LP (2005) Submission of Product Chemistry, Toxicity and Efficacy Data in Support of the Experimental Use of Cry2Ae for Use on Cotton. Transmittal of 9 Studies. 
46708901
Jansens, S.; Rouan, D. (2005) Characteristics of Cry2Ae Cotton Plants Derived From Transformation Events GHB119 and GHB714. Preliminary Report. Unpublished study prepared by Aventis Cropscience N.V. 24 p.
46708902
Habex, V. (2005) Description of the Amino Acid Sequence of the Cry2Ae Protein. Project Number: GEM2/AAS/01. Unpublished study prepared by Aventis Cropscience N.V. 5 p.
46708903
Rouquie, D. (2005) Cry2Ae (GEM2) Protein: Overall Amino Acid Sequence Homology Search with Known Toxins and Allergens. Project Number: SA/05133, TX99S005, B004973. Unpublished study prepared by Bayer Cropscience. 120 p.
46708904
Rouquie, D. (2005) Cry2Ae (GEM2) Protein: Epitome Homology and N-Glycosylation Searches. Project Number: SA/05134, TX99S006, B004974. Unpublished study prepared by Bayer Cropscience. 192 p.
46708905
Rouquie, D. (2005) Cry2Ae (GEM2) Protein: In Vitro Digestibility Study in Simulated Gastric Fluid. Project Number: SA/05032. Unpublished study prepared by Bayer Cropscience. 44 p.
46708906
Rouquie, D. (2005) Cry2Ae (GEM2) Protein: Acute Toxicity by Intravenous Injection in the Mouse. Project Number: SA/05033, B004972. Unpublished study prepared by Bayer Cropscience. 80 p.
46708907
Currier, T. (2005) Analysis to Determine if the Cry2Ae (GEM2) Protein from Cotton Leaves in Glycosylated. Project Number: BK05B003, B004964. Unpublished study prepared by Bayer Cropscience LP. 28 p.
46708908
Saey, B.; Jansens, S. (2005) PAT Cry2Ae Protein in Cotton Tissues of Events GHB119 and GHB714: Preliminary Report. Unpublished study prepared by Aventis Cropscience N.V. 9 p.
46708909
Williams, D.; Jansens, S. (2005) Preliminary Efficacy Report of Cry2Ae Cotton Plants Derived from Transformation Events GHB119 and GHB714. Unpublished study prepared by Aventis Cropscience N.V. 6 p.
47076900
Bayer Cropscience LP. (2007) Submission of Toxicity Data in Support of the Experimental Use of Bacillus thuringiensis Cry1Ab and Bacillus thuringiensis Cry2Ae for Use on Cotton. Transmittal of 2 Studies. 
47076902
Rouque, D. (2006) Cry2Ae (GEM2) Protein: Acute Toxicity by Oral Gavage in Mice. Project Number: SA/06235, BBS06/010. Unpublished study prepared by Bayer Cropscience. 62 p.
47125100
Bayer CropScience LP. (2007) Submission of Product Chemistry and Fate Data in Support of the Experimental Use of Bacillus thuringiensis Cry2Ae Insecticide Protein in Cotton. Transmittal of 2 Studies.
47125101
Jansens, S.; Rouan, D.; Habex, V. (2006) Characteristics of Cry2Ae Cotton Plants Derived from Transformation Events GHB119 and GHB714: Preliminary Report and Supplemental Information. Unpublished study prepared by Aventis Cropscience N.V. 29 p.
47125102
Rouquie, D. (2007) Cry2Ae (GEM2) Protein: In Vitro Digestibility Study in Simulated Gastric Fluid. Project Number: SA/07101, TX99L005, M/287164/01/1. Unpublished study prepared by Bayer Cropscience. 42 p.
47640000
Bayer Cropscience LP (2008) Submission of Product Chemistry, Efficacy, Toxicity, Exposure and Risk and Public Interest Data in Support of the Application for Registration of BCS TwinLink(TM) Cotton Event T304-40 x GHB 119 Cotton. Transmittal of 14 Studies.
47640001
Ferullo J. (2008) TwinLink(TM) Product Characterization - Molecular Characterization. Unpublished study prepared by Bayer BioScience N.V. 6 p.
47640002
Jesudason, P. (2008) Product Characterization - Nutritional Characterization of Twinlink. Unpublished study prepared by Bayer Cropscience LP. 8 p.
47640003
Mackie, S. (2008) Protein Expression Analysis for T304-40, GHB119 and TwinLink(TM) Transgenic Cotton. Project Number: CP08B002, BBS07/013, BB06/010. Unpublished study prepared by Bayer Cropscience LP. 155 p.
47640004
Martone, A. (2008) Analyses of Raw Agricultural Commodity (Fuzzy Seed) of TwinLink(TM) Cotton for PAT/bar, Cry2Ae and Cry 1Ab and its Non-Transgenic Counterpart for PAT/bar, Cry2Ae and Cry1Ab Proteins. Project Number: CY07B008, BTS/0066/06, BBS07/006. Unpublished study prepared by Bayer Cropscience LP. 216 p.
47640005
Martone, A. (2008) Structural and Functional Equivalence of Cry1Ab, Cry2Ae and PAT/bar Proteins Produce in Escherichia coli and Bacillus thuringiensis to the Cry1Ab, Cry2Ae and PAT/bar Proteins from Events T304-40, GHB119 and TwinLink(TM) Cotton Seed, Gossypium hirsutum. USA, 2008. Project Number: CP08B011, BTS/0077/08, BBS07/006. Unpublished study prepared by Bayer Cropscience LP. 70 p.
47640006
Martone, A. (2008) Structural and Functional Equivalence of Cry1Ab, Cry2Ae and PAT/bar Proteins Produced in Bacillus thuringiensis (Bt) and Escherichia coli to Cry1Ab, Cry2Ae and PAT/bar in TwinLink(TM) Cotton, Gossypium hirsutum, USA, 2007. Project Number: CY07B006, BBS07/013, BIOX/001/PROTSUPP/07. Unpublished study prepared by Bayer Cropscience LP. 402 p.
47640007
Bushey, D. (2008) Toxicology: Summary Human Health Assessment. Unpublished study prepared by Bayer BioScience N.V. 8 p.
47640008
Chalmers, A. (2008) Summary of Non-Target Organism Testing and Assessment of Risk of TwinLink(TM) Cotton Expressing Cry1Ab and Cry2Ae Proteins. Project Number: ECORA3. Unpublished study prepared by Bayer Cropscience LP. 17 p.
47640009
Richards, K. (2008) Evaluation of the Dietary Effect(s) of Pollen from TwinLink(TM) on Honey Bee Larvae (Apis mellifera L.) Development. Project Number: EB99X037, CAR/204/08, BTS/0081/08. Unpublished study prepared by California Agricultural Research, Inc. and Bayer CropScience Ecotoxicology Department. 48 p.
47640010
Patnaude, M. (2009) Chronic Toxicity to Collembola (Folsomia candida) Using Transgenic Lyophilized TwinLink(TM) Cotton Leaf. Project Number: EB99X036, 13798/6223. Unpublished study prepared by Springborn Smithers Laboratories. 40 p.
47640011
Stafford, J. (2008) Broiler Chicken Nutritional Equivalency Study with TwinLink(TM) Cotton. Project Number: 13798/4118, TX99X106, M/310302/01/1. Unpublished study prepared by Springborn Smithers Laboratories. 223 p.
47640012
Rinehardt, M.; Holloway, J.; Bushey, D. (2008) Agronomic Performance and Efficacy of TwinLink(TM) Cotton Obtained by Conventional Breeding of Cry1Ab Cotton Event T304-40 and Cry2Ae Cotton Event GHB119: 2007-2008 USA Production Seasons. Project Number: 07/044/102N, IR/7D/GH/MR, 07/059104N. Unpublished study prepared by Bayer Cropscience LP. 68 p.
47640013
Maclntosh, S.; Bushey, D.; Christian, M.; et al. (2008) Public Interest Document for Bacillus thuringiensis Cry1Ab and Cry2Ae Insecticidal Proteins as Expressed in Combined Trait Cotton (TwinLink(TM) Cotton) Plants. Unpublished study prepared by Bayer CropScience LP. 36 p.
47640014
Maclntosh, S.; Holloway, J.; Jansens, S.; et al. (2008) Insect Resistance Management Plan: TwinLink(TM) Cotton. Unpublished study prepared by Bayer CropScience LP. 88 p.
47641900
Bayer CropScience (2009) Submission of Product Chemistry and Toxicity Data in Support of the Application for Registration of BCS Cry2Ae Cotton Event GHB119 Cotton and the Petition for Tolerance of Bacillus thuringiensis Cry2Ae. Transmittal of 26 Studies.
47641901
Ferullo, J. (2008) Product Characterization of Cry2Ae Cotton Event GHB119. Unpublished study prepared by Bayer BioScience N.V. 24 p.
47641902
Jesudason, P. (2008) Product Characterization - Nutritional Characterization of Cry2Ae Protein. Unpublished study prepared by Bayer CropScience. 6 p.
47641903
Currier, T. (2008) Protein Expression Analysis of Cotton Event GHB119, Expressing Cry2Ae and PAT/bar Proteins. Project Number: CY07B005. Unpublished study prepared by Bayer CropScience LP. 161 p.
47641904
Martone, A. (2008) Analyses of Raw Agricultural Commodity (Fuzzy Seed) of Cry2Ae Cotton Event GHB119 for PAT/bar and Cry2Ae and its Non-Transgenic Counterpart for PAT/bar and Cry2Ae Proteins. Project Number: CY07B001. Unpublished study prepared by Bayer CropScience. 123 p.
47641905
Martone, A. (2008) Structural and Functional Equivalence of Cry2Ae and PAT/bar Proteins Produced in Bacillus thuringiensis (Bt) and Escherichia coli to Cry2Ae and PAT/bar Proteins in GHB119 Cotton, Gossypium hirsutum. Project Number CY07B002. Unpublished study prepared by Bayer CropScience. 57 p.
47641906
Capt, A.; Nennstiel, D. (2008) Toxicology (Human Health Assessment): Summary of Analysis of Open Reading Frame Homology Searches. Unpublished study prepared by Bayer Cropscience. 10 p.
47641907
Rouquie, D. (2008) Cry2Ae Protein: In Vitro Digestibility Study in Simulated Intestinal Fluid. Project Number: SA08127. Unpublished study prepared by Bayer Cropscience. 52 p.
47641908
Rouquie, D. (2008) Cry2Ae Protein with Transit Peptide Epitope Homology and N-Glycosylation Searches. Project Number: SA08209, TX99L003. Unpublished study prepared by Bayer Cropscience. 30 p.
47641909
Rouquie, D. (2008) Cry2Ae Protein with Transit Peptide Overall Amino Acid Sequence Homology Search with Known Toxins and Allergens. Project Number: SA08208, TX99L004. Unpublished study prepared by Bayer Cropscience. 444 p.
47641910
Stephenson, I. (2008) Update on Cry2Ae Protein: In Vitro Digestibility Study in Human Simulated Gastric Fluid, Overall Amino Acid Sequence Homology Search with Known Toxins and Allergens, and Epitope Homology and N-Glycosylation Searches. Unpublished study prepared by Bayer BioScience N.V. 6 p.
47641911
Rouquie, D. (2008) Cry2Ae Protein: Heat Stability Study. Project Number: SA08128, TX99L036. Unpublished study prepared by Bayer Cropscience. 49 p.
47641912
Nennstiel, D. (2008) Cry2Ae Protein: History of Safe Use. Unpublished study prepared by Bayer BioScience N.V. 7 p.
47641913
Chalmers, A. (2008) Summary of Non-Target Organism Testing and Assessment of Risk of Gossypium hirsutum: Transformation Even GHB119 Expressing Cry2Ae Protein. Project Number: ECORA2. Unpublished study prepared by Bayer CropScience LP. 50 p.
47641914
Richards, K. (2008) Evaluation of the Dietary Effect(s) of a Cry2Ae Protein on Honey Bee Larvae. Project Number: EB99L001, CAR178/07. Unpublished study prepared by California Agricultural Research, Inc. 109 p.
47641915
Patnaude, M. (2007) Laboratory Study to Determine the Effects of Cry2Ae Protein on the Predatory Beetle Coleomegilla maculata. Project Number: EB99X006, 13798/6203. Unpublished study prepared by Springborn Smithers Laboratories. 64 p.
47641916
Scott, A. (2008) Cry2Ae Protein - Toxicity to Green Lacewing (Chrysoperla rufilabris). Unpublished study prepared by Bayer CropScience LP. 4 p.
47641917
Patnaude, M. (2008) Chronic Toxicity to Collembola (Folsomia candida) using GEM2 Proteins. Project Number: EB99X0081, 13798/6204. Unpublished study prepared by Springborn Smithers Laboratories. 133 p.
47641918
Patnaude, M. (2008) Cry2Ae Protein - Acute Toxicity to Earthworms (Eisenia fetida). Project Number: 13798/6222, EB99X032. Unpublished study prepared by Springborn Smithers Laboratories. 37 p.
47641919
Sayers, L. (2008) Cry2Ae Protein - Ten Day Toxicity Test to Water Fleas (Daphnia magna) Under Static-Renewal Conditions. Project Number: 13798/6233, EB99X033. Unpublished study prepared by Springborn Smithers Laboratories. 45 p.
47641920
Martone, A. (2008) The Use of an Insect Heliothis virescens Bioassay to Determine the DT50 of the Cry2Ae Protein Produced from Bacillus thuringiensis after Aerobic Soil Degradation. Project Number: CY07B007. Unpublished study prepared by Bayer CropScience LP. 152 p.
47641921
Jesudason, P. (2008) Cry2Ae: Efficacy Assessment (Field and Laboratory). Unpublished study prepared by Bayer CropScience LP. 5 p.
47641922
Bushey, D. (2008) Request for a Waiver from the Requirement to Prepare a Public Interest Document (Cry2Ae). Unpublished study prepared by Bayer CropScience LP. 6 p.
47641923
Bushey, D. (2008) Request for a Waiver from the Requirement to Develop an Insect Resistance Management Plan (Cry2Ae). Unpublished study prepared by Bayer CropScience LP. 6 p.
47641924
Nennstiel, D. (2008) Detection Methods: rtPCR (Cry2Ae). Unpublished study prepared by Bayer BioScience N.V. 7 p.
47641925
Nennstiel, D. (2008) Protein Detection Methods: Lateral Flow Strip (Cry2Ae). Unpublished study prepared by EnviroLogix Inc. 8 p.
47641926
Nennstiel, D. (2008) Protein Detection Methods: Enzyme Linked Immuno Sorbent Assay (ELISA) (Cry2Ae). Unpublished study prepared by EnviroLogix, Inc. 10 p.
47692500
Bayer CropScience (2009) Submission of Toxicity Data in Support of the Application for Registration of BCS Cry2Ae Cotton Event GHB119 Cotton. Transmittal of 1 Study.
47692501
Patnaude, M. (2009) Cry2Ae Protein - Toxicity to Green Lacewing (Chrysoperla rufilabris) Following OPPTS Guideline 885.4340. Project Number: 13798/6229, EB99X025, BBS07/007. Unpublished study prepared by Springborn Smithers Laboratories. 65 p.
47871400
Bayer CropSciences (2009) Submission of Product Chemistry, Environmental Fate and Toxicity Data in Support of the Application for Registration of BCS Cry1Ab Cotton Event T304-40. Transmittal of 4 Studies. 
47871403
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Nontarget and Environmental Fate Studies to Test for Ecological Effects as Outlined in 40CFR 158.21-Cry1Ab. Unpublished study prepared by Bayer CropScience LP. 7 p.
47871404
Hunst, P. (2009) Bridging Data from TwinLink(TM) Cotton Avian Nutritional Study to Parental Events T304-40 and GHB119. Unpublished study prepared by Bayer CropScience LP. 7 p.
47871500
Bayer CropScience, LP (2009) Submission of Product Chemistry and Toxicity Data in Support of the Application for Registration of Cry2Ae. Transmittal of 4 Studies.
47871501
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Product Analysis Studies as Outlined in 40CFR 158.2120-Cry2Ae. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47871502
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Toxicology Studies to Test for Health Effects as Outlined in 40CFR 158.2140-Cry2Ae. Unpublished study prepared by Bayer CropScience, LP. 9 p.
47871503
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Nontarget and Environmental Fate Studies to Test for Ecological Effects as Outlined in 40CFR 158.3150-Cry2Ae. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47871504
Hunst, P. (2009) Bridging Data from TwinLink(TM) Cotton Avian Nutritional Study to Parental Events T304-40 and GHB119. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47871600
Bayer CropScience, LP (2009) Submission of Product Chemistry and Toxicity Data in Support of the Application for Registration of BCS TwinLink(TM) Cotton Event T304-40xGHB 119 Cotton. Transmittal of 6 Studies. 
47871601
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Product Analysis Studies. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47871602
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Toxicology Studies to Test for Health Effects as Outlined in TwinLink. Unpublished study prepared by Bayer CropScience. 9 p.
47871603
Coats, I. (2009) Request for Waivers from the Requirement to Conduct Certain Nontarget and Environmental Fate Studies to Test for Ecological Effects - TwinLink. Unpublished studies prepared by Bayer CropScience. 7 p.
47871604
Currier, T. (2009) Protein Expression Analysis for Pat/bar, Cry1Ab and Cry2Ae Proteins in Whole TwinLink(TM) Plants Grown in the Field. Project Number: CP09Q005. Unpublished study prepared by Bayer CropScience, LP. 28 p.
47871605
Hunst, P. (2009) Investigation of Synergistic Effects of the Cry1Ab and Cry2Ae Proteins expressed in TwinLink(TM) Cotton (BCS-GH004-7 x BCS-GH005-8) on Target Insect Pests Using Leaf Bioassay. Unpublished study prepared by Bayer CropScience, LP. 8 p.
47871606
Coats, I. (2009) Supplement to MRID # 47640014: Insect Resistance Management Plan: TwinLink(TM) Cotton. Unpublished study prepared by Bayer CropScience, LP. 7 p.
47903100
Bayer CropScience LP (2009) Submission of Toxicity Data in Support of the Application for Registration of BCS TwinLink(TM) Cotton Event T304-40 x GHB 119 Cotton. Transmittal of 1 Study.
47903101
Hunst, P. (2009) Request for Waiver from the Requirement to Conduct an Avian Oral Toxicity Test: TwinLink. Unpublished study prepared by Bayer CropScience LP. 16 p.
47903300
Bayer CropScience LP (2009) Submission of Toxicity Data in Support of the Application for Registration of BCS CRY2AE Cotton Event GHB119. Transmittal of 1 Study.
47903301
Hunst, P. (2009) Request for Waiver from the Requirement to Conduct an Avian Oral Toxicity Test: TwinLink. Unpublished study prepared by Bayer CropScience LP. 16 p.
47928800
Bayer CropScience, LP-BioScience. (2009) Submission of Efficacy Data in Support of the Registration of BSC TwinLink(TM) Cotton Event T304-40 X GHB 119 Cotton. Transmittal of 1 Study.
47928801
Coats, I. (2009) Addendum to Product Chemistry for TwinLink: Structural and Functional Equivalence of Cry1Ab and Cry2Ae and PAT/bar Proteins Produced in Bacillus thuringiensis (Bt) and Escherichia coli to Cry1Ab, Cry2Aa and PAT/Bar in TwinLink(TM) Cotton, Gossypium hirsutum, USA, 2007: Final Report. Project Number: CY07B006. Unpublished study prepared by Bayer CropScience, LP. 7 p.
48177500
Bayer CropScience LP (2010) Submission of Toxicity Data in Support of the Applications for Registration of BCS TwinLink(TM) Cotton Event T304-40 X GHB 119 Cotton, BCS Cry 1Ab Cotton Event T304-40, and BCS Cry2Ae Cotton Event GHB119 and the Petition for Tolerance of Cry2Ae. Transmittal of 1 Study.
48177501
Stafford, J. (2010) Northern Bobwhite (Colinus virginianus) Dietary Toxicity Test (LC50) with GHB119 (Cry2AE) Cottonseed Meal. Project Number: 13798/4130, TX44L001. Unpublished study prepared by Springborn Smithers Laboratories. 60 p.
48427900
Bayer CropScience (2011) Submission of Efficacy Data in Support of the Experimental Use of T304-40 x GHB119 x COT102 x COT67B in Cotton. Transmittal of 1 Study.
48427901
Robinson, T. (2011) Cry1Ab, Cry2Ae, PAT/bar, FLCry1Ab, Vip3Aa, APH4 Levels in Selected Tissues of the Combined Trait Product T304-40 x GHB119 x COT102 x COT67B (BCS-GHOO4-7 x BCS- GH005-8 x SYN-IR1O2-7 x SYN-IRO67-1). Project Number: 10/BM45L001. Unpublished study prepared by Bayer CropScience, LP. 22 p.
48471900
Bayer CropScience (2011) Submission of Product Chemistry Data in Support of the Experimental Use of Cry2Ae (GEM2) in Cotton. Transmittal of 1 Study.
48471901
Currier, T. (2011) Analysis to Determine if the Cry2Ae (GEM2) Protein from Cotton Leaves is Glycosylated: Final Report. Project Number: BK05B003. Unpublished study prepared by Bayer CropScience. 32 p.
48480000
Bayer CropScience LP (2011) Submission of Product Chemistry and Efficacy Data in Support of the Applications for Registration of BCS Cry1Ab Cotton Event T304-40, BCS Cry2Ae Cotton Event GHB119 and BCS TwinLink(TM) Cotton Event T304-40 x GHB 119 Cotton, and the Petition for Tolerance of Cry2Ae. Transmittal of 11 Studies.
48480001
Habex, V.; Verhaeghe, S. (2008) Detailed Insert Characterization of Gossypium hirsutum Transformation Event GHB119 by Southern Blot Analysis. Project Number: BBS07/009, M/308404/02/1. Unpublished study prepared by Bayer BioScience N.V. 43 p.
48480002
Moens, S. (2008) Confirmation of the Absence of Vector Backbone Sequences in Gossypium hirsutum Transformation Event GHB119. Project Number: BBS06/007, M/311127/02/1. Unpublished study prepared by Bayer BioScience N.V. 40 p.
48480003
Moens, S.; Criel, I. (2008) Detailed Insert Characterization of Gossypium hirsutum Transformation Event T304-40. Project Number: BBS07/014, M/311245/03/1. Unpublished study prepared by Bayer BioScience N.V. 62 p.
48480004
Habex, V.; Van Herck, H. (2007) Confirmation of the Absence/Presence of Vector Backbone Sequences in Gossypium hirsutum Transformation Event T304-40: Final Report. Project Number: BBS05/003, M/308400/02/1. Unpublished study prepared by Bayer BioScience N.V. 41 p.
48480005
Currier, T. (2011) Characterization of Cry1Ab and (Inert Ingredient) Proteins Produced in Event T304-40 Cotton, Gossypium hirsutum USA, 2011. Project Number: 10/BM99L508, M/406631/01/1. Unpublished study prepared by Bayer CropScience LP. 31 p.
48480006
Moens, S. (2011) Characterization of Cry2Ae and (Inert Ingredient) Proteins Expressed in Gossypium hirsutum Transformation Event GHB119. Project Number: BIO2/005/PROTEQUL/512, M/406615/02/1. Unpublished study prepared by Bayer BioScience N.V. 35 p.
48480007
Currier, T. (2011) Structural Stability Analysis of Gossypium hirsutum Combined Event TwinLink(TM) X GlyTol (GHB614): Final Report. Project Number: CP08Q012/2. Unpublished study prepared by Bayer CropScience LP. 34 p.
48480008
Currier, T.; Harbin, A. (2011) Protein Expression Analysis for GlyTol x TwinLink(TM) (GHB614 x T304-40 x GHB119) Cotton Grown in the Field, USA 2010. Project Number: 10BM99L440B, M/406245/01/1. Unpublished study prepared by Bayer CropScience LP. 125 p.
48480009
Bushey, D.; Jansens, S.; Becker, A. (2010) Insect Bioassay of Plant Material Expressing Cotton Events T304-40, GHB119, and TwinLink(TM) (T304-40xGHB119). Project Number: BIO2/005, INSECT/BIOASSAY/FROM/MONHEIM. Unpublished study prepared by Bayer CropScience. 22 p.
48480010
Mullins, W. (2011) High Dose Study: Estimating the Mortality of Heliothis virescens in a Mississippi Cotton Field on Caged Plants of T304-40 x GHB119 (TwinLink) Cotton (BCS-GHOO4-7 x BCS-GHOO5-8) and Parental Events T304-40 (Cry1Ab) and GHB119 (Cry2Ae). Unpublished study prepared by Bayer CropScience LP. 16 p.
48480011
Dennehy, T. (2011) High Dose Study: Estimating the Mortality of Heliothis virescens in a Greenhouse on Plants of T304-40 x GHB119 (TwinLink) Cotton (BCS-GHOO4-7 x BCS-GHOO5-8) and Parental Event Plants T304-40 (Cry1Ab) and GHB119 (Cry2Ae). Project Number: BA44Y001. Unpublished study prepared by Bayer CropScience LP. 16 p.
48495100
Bayer CropScience (2011) Submission of Efficacy Data in Support of the Application for Registration of the TwinLink. Transmittal of 1 Study.
48495101
Dennehy, T. (2011) A Risk Assessment of TwinLink(TM) Cotton and Tobacco Budworm (Heliothis virescens) in the Midsouth. Unpublished study prepared by Mississippi State University. 15 p.

Other References

Safe use  -  mode of operation

Ferré, J. and Van Rie, J. 2002. Biochemistry and genetics of insect resistance to Bacillus
thuringiensis. Ann. Rev. Entomol. 47:501-533.

Organization for Economic Co-operation and Development (OECD). 2007. Consensus document
on safety information on transgenic plants expressing Bacillus thuringiensis-derived
insect control protein. 109 pages.




